The Scottish Government

« Previous | Contents | Next »

Listen

Drug Treatment and Testing Orders: Evaluation of the Scottish Pilots

CHAPTER SEVEN: THE COSTS OF DTTOS

INTRODUCTION

7.1 7.1 Data presented elsewhere in this report provides an early indication that DTTOs can have a positive impact upon both continued offending and drug misusing behaviour. However, consideration of the benefits of DTTOs must also include consideration of the costs involved. In this chapter both the direct and indirect costs of an average DTTO during the pilot period are presented. These costs have been compared with the estimated costs to the community of drug misuse and drug-related crime and estimates of savings achieved through a variety of treatment modalities.

THE COST OF DRUG MISUSE

7.2 The cost to society of drug misuse is substantial. The Scottish Executive's Drugs Action Plan: Tackling Drugs in Scotland: Protecting our future (2000), estimates the total annual cost to the public purse in Scotland at 250 million, although this estimate does not include the cost of absenteeism, crime and the victims of crime. In addition, the report cites research in Glasgow estimating the cost of drug-related theft at between 200 and 300 million.

7.3 The Government's Ten-Year Strategy: Tackling Drugs to Build a Better Britain (1998) estimated that spending across the UK was over 1 billion with the actual cost to society estimated at over 4 billion, of which 1 billion was incurred within the criminal justice system through drug-related crime.

7.4 In the USA, a study prepared for the National Institute on Drug Abuse (NIDA) by the Lewin Group (1992) estimated the economic cost of drug misuse at $98 billion; with the majority of these costs (46%) being borne by the public purse. Only $4.4 billion (4.5%) of this sum represented expenditure on specialist treatment services. Of these costs, over $59 billion (60%) were attributable to drug related crime with a threefold increase in crimes attributed to drug misuse over data collected in a similar study in 1985. The same study estimated that drug misuse contributed to 25 - 30% of income-generating crime. Held (1998) suggests a figure of $24 billion in costs to the victims of drug related crime "excluding the value of pain, suffering and lost quality of life".

ESTIMATING THE COST EFFECTIVENESS OF DRUG TREATMENT INTERVENTIONS

7.5 Although published economic evaluations of substance misuse treatment interventions are not new - with Swint and Nelson (1977), Hertzman and Montague (1977) and others demonstrating the applicability of economic analysis to the performance of early specialist alcohol treatment units over twenty years ago - they are relatively rare.

7.6 In part, this is a result of the complexity of the task. Studies of this type in the broader health-care arena have tended to estimate intervention benefits in terms of reduced health-care costs and working days lost (Werthamer, 1998). However, participants in drug misuse treatment interventions are most likely to be young people, dislocated from the legitimate labour market with an extremely variable take-up of non drug misuse-related health care.

7.7 Moreover, many of the benefits cited by the proponents of such treatment services - improved self esteem, reductions in familial conflict, improved mental health - are difficult, if not impossible, to assign a monetary value to.

7.8 Most published examinations of health-related interventions are either cost effectiveness or cost benefit analyses. Cost effectiveness analyses assume that the programme evaluated and the available alternative options will produce broadly similar outcomes (Werthamer, 1998). The value of the outcomes themselves is not considered; rather the cost of producing them. Cost benefit analyses are traditionally used to assess whether an intervention is worthwhile in and of itself. No comparison is made to other programmes. Cost benefit analyses are normally designed to measure whether the benefits of an intervention outweigh its costs and justify the necessary allocation of resources.

7.9 In the USA, a number of studies have estimated the cost benefits of both community based ambulatory treatment (Harwood et al., 1988; Cartwright and Kaple, 1991; Bradley, French and Rachal, 1994) and in-patient, or residential, provision (French et al., 1999; De Leon, 2001; Rawlings, 2001). Harwood et al. (1988) significantly note that treatment episodes can result in reduced levels of criminal activity even where employment prospects and earnings have shown no significant improvement.

7.10 A number of studies (French et al., 1993; French et al., 1999; Cartwright and Kaple, 1991), show a significant correlation between length of time in treatment and successful outcome, whilst Project Match (1996), one of the most comprehensive studies ever to be undertaken into treatment efficacy, noted the importance of external support in reducing both substance misuse and criminal activity. Gerstein and Harwood (1990) conclude that even apparently expensive treatment interventions are cost effective over a relatively short period of time.

7.11 In the UK, the paucity of studies evaluating the effectiveness of treatment services for drug misuse was noted by the Department of Health Task Force (1996) although a number of studies of alcohol treatment provision (McCrady et al., 1986; Holder, 1987; Holder et al., 1991; Goodman et al., 1992) have indicated positive cost benefits.

7.12 The National Treatment Outcome Research Study (NTORS) examined outcomes for 1075 UK clients recruited into both residential and community treatment modalities. At the one-year review, the study estimated that savings of at least 5 million had been achieved across the whole cohort. On the basis of these findings, the NTORS team concluded that, "for every extra 1 spent on drug misuse treatment, there is a return of more than 3 in terms of costs savings associated with lower levels of victim costs of crime and reduced demands upon the criminal justice system. The total costs savings to society may be even greater than this" (NTORS, 1996).

ESTIMATING THE DIRECT COST OF THE PILOT DTTOS

7.13 An in-depth cost effectiveness or cost benefit analysis comparing the two sites was not feasible. This was due to the different methodologies adopted within the two pilot areas, with Fife opting to provide the bulk of its treatment resources 'in-house' and Glasgow largely relying upon the purchasing of external contractors. As a result, whilst overall outcome and delivery costs can be estimated and compared, some operational management costs were particular to the individual provider.

7.14 The cost analysis provided in this section is based upon the direct costs for both pilot teams (salaries, travel, accommodation etc.) plus the cost of any externally purchased services (treatment interventions, urine analysis etc.). Other, indirect costs, are considered in a subsequent section.

7.15 In addition, even though there is a strong indication that, in most cases, a custodial disposal would have been the likely outcome had it not been for the availability of a DTTO, a number of other alternative disposal options were available and could have been effected (and in some of these cases a community-based disposal may have been breached and a custodial sentence imposed). We describe later in this chapter the assumptions we have made about the sentences that DTTOs had replaced.

7.16 Financial statistics provided related to the first year of operation with all that this entailed in respect of marketing a new service, recruiting staff, resolving 'teething problems' and rationalising throughput. In the case of Fife, the period covered is July 2000 to mid-April 2001; slightly less than a full year. Costs provided by the two pilot areas covered the financial year 2000/2001.

7.17 Since the two pilots recorded cost information under different headings, we have simplified these data in Table 7.1 by identifying staff costs, office and travel costs and purchase of other services. This last item includes provision of treatment services, prescribing and drug testing.

Table 7.1: DTTO scheme costs

* Fife costs based upon budget estimates provided.

7.18 Using these figures, a mean cost per order was estimated for both pilots by establishing the mean cost per month per order. The breakdown of the length of orders provided elsewhere in this report (Table 3.3) indicates that a total of 732 DTTO months were provided in Glasgow and 822 in Fife. Dividing the total expenditure by the number of DTTO months offered in each pilot provided a mean cost per month of an order of 503 in Glasgow and 487 in Fife (Table 7.2). This figure can be scaled up to provide an average cost for orders of varying lengths (Table 7.3) and a mean cost per order based on the proportionate use of orders of different lengths in the pilot period (see Table 3.3). This produces an average direct cost per DTTO of 8,108 across the two pilot schemes.

Table 7.2: Cost per DTTO month (s)

Table 7.3: Direct costs (s) of orders of varying lengths

ESTIMATING THE INDIRECT AND TOTAL COSTS OF THE PILOT DTTOS

7.19 The indirect costs associated with the provision of DTTOs are more complex to calculate. Some costs such as the cost of child-care arrangements and travel for clients attending treatment services are not particular to this client group and might reasonably be expected to feature as indirect costs in respect of other non-custodial disposals.

7.20 In addition to the direct costs of assessment, supervision and treatment must be added the costs of monthly reviews. In the absence of any published information regarding the cost of DTTO review hearings, we have assumed that the costs of reviews would be similar to those for a plea at first diet in the sheriff court under summary proceedings. Based on figures published by the Scottish Executive for the year 1999-2000, this produces a court cost of 62 per review hearing. Adding this to the direct costs on the assumption that each order is reviewed monthly (and that there are no additional costs associated with the representation of offenders by defence agents at the reviews) produces the total costs illustrated in Table 7.4 and an average total cost per DTTO across the two sites of 9,124. Whilst these costs appear high, they pertain to an intensive, demanding and relatively lengthy order. On a pro rata basis they compare favourably with the estimated roll-out costs for Restriction of Liberty Orders, at 4,560 for a six-month order (Lobley and Smith, 2000).

Table 7.4: Total costs (s) of orders of varying lengths

7.21 It should be noted that the cost estimates provided above do not make allowance for all offenders referred to the system and processed by the DTTO staff. During the study period, Glasgow received a total of 155 referrals and Fife, 114. The cost of each DTTO will therefore be slightly less than the above estimate, which assumes that all the pilot schemes' costs are spread across the orders that are actually imposed. On the other hand, there will be additional costs associated with breached orders (DTTO staff costs, further court costs and the costs of any alternative sentences imposed) which will have the net effect of increasing the unit cost of a DTTO. We have not included costs of breached orders in this analysis since the low number of breaches during the period covered by the evaluation makes the likely longer term breach rate and the sentences imposed on revocation difficult to assess.

7.22 However, these costs should not be seen as an indication of probable ongoing costs where DTTO services are provided on a permanent basis. The mean cost per order provided here is likely to be a significant overestimate of ongoing unit costs based upon a pilot period of high expenditure and low throughput.

7.23 The average cost of 9,124 is based upon an average of 48 orders across the two pilot schemes. The number of orders that could be supervised would be higher once the pilot schemes had become established and this would result in a reduction in unit costs. For example, if it was assumed that 60 orders per annum could be supervised by each scheme, this would produce a revised unit cost of 7,293 per average-length DTTO.

ESTIMATING THE COST IMPLICATIONS OF A ROLL-OUT OF ORDERS

7.24 When considering the costs of DTTOs it is important also to take into account the likely costs of the alternative sentences that the orders might have replaced. In Chapter Three it was found that 38 per cent of offenders who were assessed for a DTTO, but who received an alternative disposal, were imprisoned, 23 per cent received a probation order with a drug treatment requirement, 15 per cent received a probation order and 24 per cent received other disposals. As we have noted elsewhere in this report, it appeared that DTTOs were generally offered to high tariff offenders. It is assumed, therefore, that in the absence of DTTOs, offenders would have received a custodial sentence or probation with a condition of treatment. Of the 58 offenders assessed as unsuitable for DTTO and who were given an alternative disposal of custody (36) or probation with treatment (22), 62% were given custody and 38% probation with a condition of treatment (see Table 3.2). These percentages have been used to calculate the likely costs of alternatives to DTTOs.

7.25 Data published by the Scottish Executive estimate the cost of six-months in prison as being 14,057 for 1999/2000 (Scottish Executive, 2002). We have no basis on which to estimate accurately the proportionate costs of longer or shorter sentences. We therefore assume, for the purposes of these estimates, that the cost of imprisonment per month is similar (at 2343 per month) for a 12-month sentence (i.e. six months in prison) and for a shorter sentence. There are no published costs for a probation order with a drug treatment requirement, but we have assumed that this will be lower than the mean cost of a DTTO since it will exclude reviews. On this basis we estimate the average cost of a probation order with a drug treatment requirement as being approximately 6,500.

7.26 Based on our revised estimate of a DTTO in an established scheme (7,293), the average annual cost of providing 60 DTTOs would be in the region of 437,580 (including the costs of review hearings). If it were assumed that a DTTO replaces an average prison sentence of six months (i.e. three months served) in 62 per cent of cases and a probation order with a drug treatment requirement in 38 per cent of cases, the costs of the alternative sentences would be 261,479 and 148,200 respectively (and 409,679 in total). If DTTOs replaced prison sentences of nine-months on average, the cost of these alternative sentences would be 540,400. Using these estimates, the average cost of a DTTO compares favourably with the cost of other disposals. This, combined with the significant savings that are expected to accrue from a reduction in drug use and associated crime, suggests that DTTOs are a cost-effective addition to the repertoire of sentences available to the courts.

ESTIMATING THE COST IMPACT OF DTTOS ON DRUG-RELATED OFFENDING

7.27 Elsewhere in this report, we have noted significant reductions in spending upon illicit drugs amongst those given a DTTO. Self-report evidence from offenders at the six-month review stage suggested that spending on drugs had reduced to 57 per week from 490 per week prior to the order.

7.28 Edmunds et al. (1998), in a study which compares the cost findings of a total of seven recent UK studies, estimate that the mean average individual spend on illicit drugs is 200 per week: considerably less that the 490 per week reported by offenders in this study. Using this figure and multiplying by a factor of 3 (the multiplier used in the Government's Tackling Drugs Together [1995] to reflect the low illicit resale value of stolen goods), the authors estimate a mean individual cost, in stolen property alone, of 31,200 per annum.

7.29 Applying this formula to the 95 offenders in this study, results in an estimated reduction of costs (in stolen property) from 76,440 per annum at the outset of the order to 8,892 per annum at the six monthly review: an average saving of 33,774 per offender over the period. Even if only 50% of these savings could be achieved, this would still represent savings to the community of over 3 million per annum: more than three times the cost of funding the two pilots.

« Previous | Contents | Next »