Money for medication: financial incentives to improve medication adherence in assertive outreach

AIMS AND METHOD

Although financial incentives to improve treatment adherencehave been found effective in various medical specialties, thereare few systematic data on their use, ethical background andeffectiveness in psychiatry. We explored the practice of andpossible ethical problems associated with direct financial incentivesto improve adherence in assertive outreach teams in England.We also report clinical observations of a money for medicationscheme with five assertive outreach patients in East London.

RESULTS

None of the assertive outreach teams that responded to the survey(response rate 47%) uses financial incentives. Attitudes ofteam managers towards the practice were mostly negative, oftenregarding it as unethical. Specific concerns were related topossible coercion and a negative impact on the therapeuticrelationship. Out of five patients studied, four accepted the offer of money and had improved adherence; three remained withouthospital admission since entering the scheme.

CLINICAL IMPLICATIONS

Money for medication might be a non-coercive and effective optionto achieve medication adherence in otherwise non-adherent assertiveoutreach patients. However, ethical issues need further explorationand controlled trials are required to establish the effectiveness.

Introduction

TOP

Introduction

Lack of adherence to antipsychotic medication has been identifiedas a major obstacle to effective treatment in schizophrenia (Leucht et al, 2003). Non-adherence rates are estimated tobe 20–50% in patients in general adult psychiatric services(Nose et al, 2003). This figure is likely to be substantiallyhigher in assertive outreach patients, as non-adherence isone of the possible criteria for referral (Priebe et al, 2003;Weaver et al, 2003).

Various clinical interventions have been tried to achieve adherencein non-adherent patients. These include ‘compliance’therapy (O’Donnell et al, 2003), psychotherapy, familyeducation, telephone prompting and psychoeducation. A reviewby McDonald et al (2002), focusing on studies of chronic healthproblems, found only modest effects of all described interventions(mainly information, reminders, reinforcement and family therapy).A recent meta-analysis (Nose et al, 2003) of studies to enhanceadherence in psychiatric patients concluded that some interventionsare effective (with an odds ratio >2 compared with usualcare).

Giuffrida & Torgerson (1997) found favourable results in10 out of 11 randomised controlled studies of financial incentivesto enhance adherence to anti-tuberculotic drugs, dental care,weight reduction and cocaine abstinence programmes, and antihypertensivetreatment with odds ratios of up to 7 for antituberculosistreatment (Bock et al, 2001; Tulsky et al, 2004). One non-randomisedstudy (Carey & Carey, 1990) of patients with dual diagnosisfound that modest rewards enhanced attendance. All studieswere from the USA and the incentives ranged from $5 to $1000for a completed programme or single intervention.

In psychiatric care, there are few data on how financial incentivesare used to improve adherence and practically no research evidenceon their effectiveness. In the USA, social services frequentlycontrol the payment of social security benefits and link thisto care adherence (Elbogen et al, 2005), often within a frameworkof ’mandated’ community treatment (Monahan et al,2001, 2005). The practice has been criticised (Cogswell, 2004), but has not been systematically researched. In the UK, moneyincentives have been used to assure that patients undergoingclozapine treatment attend the clinic for blood count controls(Pereira et al, 1999). In research settings, patients have acknowledged that the offer of money would increase their willingnessto participate in medication trials (Roberts et al, 2003).

This study investigates the use of direct financial incentivesto increase medication adherence in assertive outreach teamsin England, the team managers’ attitudes to direct financialincentives, and how can they be addressed, and the effectiveuse of financial incentives to increase adherence in assertiveoutreach patients.

Method

TOP

Method

We sent questionnaires to team managers of 150 assertive outreachteams in England asking whether financial incentives were usedto increase medication adherence, what the attitudes to sucha practice are, and whether the issue had been discussed withinthe team. After 3 weeks, teams were reminded via the telephone.Results were analysed descriptively.

We also offered financial incentives (£5–15 paymentfor single depot injections) to five formerly non-adherentassertive outreach patients in East London. The team has anoperational policy for the use of money for medication. Inclusioncriteria were hospitalisations with a total duration of atleast 50 days in the preceding 2 years, which according to clinical judgement were the result of a lack of medication adherence,and a failure of all other available therapeutic strategiesto achieve adherence. The amount of money offered dependedon the frequency of depot injections. Experiences were recorded.Patients’ characteristics, including hospital admissionsprior to and following the introduction of financial incentives,were obtained from the medical files.

Results

TOP

Results

The questionnaire was returned by 70 assertive outreach teams(response rate 47%). No team had used financial incentivesover the past 2 years. Seven (10%) team managers reported thatthey had used food and other indirect incentives to help withengagement, but not as a direct reward for adherence to medication.No team has an operational policy governing the possible useof financial incentives.

Only 12 team managers (17%) had no objections to the use offinancial incentives, whereas 53 managers (76%) specified objectionsto such a practice. Five team managers (7%), did not give theiropinion. Objections mentioned by managers are listed in Table 1.

View this table:
[in this window]
[in a new window]

Table 1. Reasons given by assertive outreach teams for objections to the use of financial incentives (n=53)

Apart from general concerns about financial incentives beingunethical (42%), specific issues were raised regarding thepossibly coercive nature of this treatment (8%) and a potentialnegative impact on the therapeutic relationship (9%).

In 43 teams (61%), the issue had not been discussed. In 23 teams(33%) the issue had been discussed, and in 6 of these teamsthe majority of team members had clearly voted against it.Four questionnaires (6%) did not indicate whether the issuehad been discussed.

From summer 2003 onwards, financial incentives were offeredto five assertive outreach patients in East London. One patientdeclined the offer and still remains difficult to engage. Diagnosesof the other four patients as well as number of hospitalisationssince onset of illness, days in hospital in the 2 years priorto the use of financial incentives and days in hospital sinceentering the scheme are shown in Table 2.

View this table:
[in this window]
[in a new window]

Table 2. Characteristics of four assertive outreach patients participating in the financial incentive scheme in Newham

Three patients were not hospitalised after the introductionof financial incentives. The fourth patient had entered thescheme only a few months before this analysis and had alreadybeen readmitted, although adherence had improved.

Clinical experiences of the team with the new intervention arerather favourable. All four patients on the scheme were ableto retain their independent accommodation and had fewer problemswith their neighbours and the police than before. So far, noother patients have complained about unequal treatment and/orhave demanded to be paid for taking their medication. One of the patients on the scheme had asked for the incentive to beincreased. This was declined and he has remained on the schemewith the original incentive. We have not yet tried to terminatethe use of financial incentives in any of the patients, andcannot comment on problems that might arise.

Discussion

TOP

Discussion

Our survey was restricted to assertive outreach teams and theresponse rate was less than 50%. Financial incentives mightbe used in assertive outreach teams that did not respond orin community mental health or early intervention teams. However,as our survey did not reveal the use of money for medication by a single team, wide-spread use of this approach is unlikely.Some teams use food and other non-financial incentives to enhanceadherence in a more covert way.

The attitudes of assertive outreach team managers towards usingdirect financial incentives to improve adherence were rathernegative, and many regarded the practice as unethical. Ethicalconcerns have been raised in the literature with respect tothe application of depot medication (Roberts & Geppert, 2004) and assertive outreach itself (Watts & Priebe, 2002; Williamson, 2002).The use of financial incentives adds further ethical issues to those general concerns. The fact that financial incentiveshave been effective in improving treatment adherence in othermedical specialties is important, but does not fully addressthe ethical problems in psychiatry.

Beauchamp (2003) describes four categories for judging theethical dimension of medical interventions:

beneficence (the obligation to provide benefits and balancebenefits against risks)
non-maleficence (the obligation toavoid the causation of harm)
respect for autonomy (the obligationto respect the decision-making capacities of autonomous persons)
justice (obligation of fairness in the distribution of benefitsand risks).

One of the more specific concerns regards a possible coercivenature of financial incentives and with it a negative impacton the autonomy of the service user. Coercion is defined as‘the use of threats or orders to make someone do somethingthey do not want to do’ (Longman, 2003). Wertheimer (1993)proposed the following practical approach to decide whethera (therapeutic) proposal is coercive or not:

‘The standard view is that threats coerce but offers donot. And the crux of the distinction between threats and offersis that A makes a threat when B will be worse off than in somerelevant baseline position if B does not accept A’s proposal,but that A makes an offer when B will be no worse off thanin some relevant baseline position if B does not accept A’s proposal.’

According to these definitions, money for medication is notcoercive, as it adds only a small amount of money (comparedwith the average benefits) to what the service user alreadyreceives, much in the sense of a reward in the framework ofbehavioural reinforcement. Financial incentive does not affect any entitlement to benefits or any other component of care.In Wertheimer’s definition, it can be seen as an offer,not a threat.

Considering Beauchamp’s criteria with our small samplein mind, this approach appears to be beneficial in terms ofreduced hospital admissions, there is no harm intended or caused,the service user can revoke the agreement at any time withoutnegative effect (other than having £5–£15less) and the treatment is offered to all service users fulfillingthe criteria. This is clearly different from practice in the USA, where not only social benefits but entitlement to housingprovision and also the freedom to stay in the community (asopposed to involuntary hospital treatment) are tied to adherenceto treatment programmes and medication via a payee system,tenancy contracts or so-called out-patient commitment orderedby a mental health court (Monahan et al, 2001; Swartz & Monahan, 2001).

A further concern raised by team managers regards the possiblenegative impact on the therapeutic relationship. In our clinicalpractice we have not observed this. One of the reported patientson the scheme in East London now sits, as a service user representative,regularly on interview panels for new team staff. In practice,the findings of our clinical observations in East London arerather encouraging. At least three out of five patients whowere offered the scheme showed a remarkable improvement. Ofcourse ethical issues are paramount, and if regarded as unethical,money for medication should not be used, as beneficence isonly one of the criteria to assess ethics.

In addition to ethical issues there are also practical questionsthat need to be addressed. For instance, how is mental capacityto agree to the scheme assessed? Where does the money for thepatients come from, particularly if the practice is used ona larger scale? Will the money received for medication havean impact on social and disability allowances? What is the appropriate amount of money? Can financial incentives also be used to improveadherence to oral medication or only depot injections? Whenand how should the scheme be discontinued? Answers to thesequestions may be contentious and difficult. However, in ourview this is no reason to prevent both an open debate on ethicaland practical issues related to money for medication and studiesto elicit more evidence that can inform the debate. Financialincentives might be a treatment option for a high-risk groupof non-adherent patients with whom all other interventionsto achieve adherence have failed.

Acknowledgments

We thank all managers of assertive outreach teams who returnedthe questionnaires and the User Advisory Group on Researchof the East London and City Mental Health Trust for providinghelpful suggestions.

References

TOP