A pharmacoeconomic analysis of the impact of therapeutic drug monitoring in adult patients with generalized tonic-clonic epilepsy

Abstract

Aims

To carry out a retrospective pharmacoeconomic analysis of the impact of therapeutic drug monitoring (TDM) in adult patients with generalized tonic-clonic epilepsy in an academic, non profit making organization.

Methods

Twenty-five patients who had undergone TDM were compared with 25 age, disease and duration of drug therapy matched controls who had not undergone TDM. Only direct costs were calculated. These included cost to the hospital of providing the TDM service, cost to the hospital per seizure saved, and cost to the patient per seizure saved.

Results

Patients undergoing TDM had much more effective seizure control (P = 0.00032, OR 4.846, 95% confidence interval 1.29,18.3), fewer adverse events, better earning and were more likely to be married than the control group.

Conclusions

In patients with adult onset epilepsy, a minimum of two drug estimations per year offers significant benefit in terms of better seizure control, fewer adverse events and greater chances of remission.

Introduction

Monitoring of plasma concentrations of anticonvulsants is used to enhance the efficacy of therapy by revealing unrecognized under or over dosage, detecting failure of compliance, drug–drug interactions or indicating when there is little point in persisting with a particular treatment [1]. Over the last 30 years, with the lessons learnt from the use of therapeutic drug monitoring (TDM) and growing concern for savings in health care costs, the principles of pharmacoeconomics have begun to be applied to various fields including TDM [2].

The TDM clinic at the Department of Clinical Pharmacology receives approximately 1800 old cases and 1500 new cases yearly, which are referred for noncompliance, toxicity, treatment failure, or drug–drug interactions including with traditional systems of medicine [3–5]. Patients at this TDM clinic (who belong to the lower socio-economic strata) are charged only the cost of consumables since ours is an academic, nonprofit making institution. We attempted to assess the impact of this TDM service in adult patients of generalized tonic-clonic (GTC) epilepsy by carrying out a retrospective, post hoc pharmacoeconomic analysis of a cohort of 25 patients who had undergone TDM and compared them with 25 controls who were matched for age, disease, duration of drug therapy and duration of epilepsy clinic attendance.

Methods

Subjects

Patients who satisfied inclusion-exclusion criteria were divided into two groups (n = 25 each). Group A patients were selected from the TDM clinic and had undergone TDM while Group B were selected from the epilepsy clinic who had simultaneous attendance and had not undergone TDM. For group A, 50 patient records were screened and patients contacted by mailing a letter. The first 25/50 who responded and consented were selected for the study. Similarly, the first 25 patients fulfilling the inclusion and exclusion criteria were selected from the epilepsy clinic.

Inclusion was those who had onset of epilepsy between 18 and 35 years, epilepsy duration of at least 4 years, age at the time of interview between 24 and 40 years, duration of antiepileptic medication at least 4 years, a minimum of two TDMs done per year (group A only), at least two visits to the epilepsy clinic/year for at least 2 years (both groups). Excluded were those with secondary epilepsy, seizure type other than GTC, concomitant diseases or concomitant drug therapy.

Study design

All patients were interviewed using a pre-designed questionnaire that consisted of questions about the age at onset of epilepsy, nature of antiepileptic medication past and present, years of attendance at the TDM clinic (Group A only), years of attendance at the epilepsy clinic (both groups), number of years without TDM (Group B only), number of TDMs done so far (Group A only), seizure control before and after entry into the TDM and epilepsy clinics, delay in start of antiepileptic medication, adverse events past and ongoing at interview time (based on history and neurological examination), earning status past and at the time of interview, and marital status. Only direct costs were calculated from the perspective of a large public hospital in a developing country.

Cost calculations

a) Cost to the hospital of providing the TDM service was calculated by summing the cost of equipment, depreciation, repairs and maintenance, interest on capital, cost of consumables and salaries of the staff per year. b) The cost per sample analysed to the hospital was calculated by dividing the total cost of providing the TDM service per year by the total number of TDM samples analysed in that year. (c) Savings as a result of seizure control were calculated as follows:

Cost to the hospital per seizure saved=Cost per TDM sample to the hospital (Rs 147)×number of TDM done for these 25 patients per year (286).

Number of seizures saved per year (3224) = 13.03 rupees.

Cost to the patient per seizure saved (n = 25) =TDM cost charged to the patient (Rs 30) × number of TDMs done per year (286).

Number of seizures saved per year in the patient (3224) = 2.67 rupees.

Statistical analysis

Between group comparison for period before TDM (Group A) or 2 years prior to interview (Group B) was done to assess if the two groups were comparable at the outset. Within group comparison for Group A was done pre and post TDM referral. Likewise, for group B, two observation points 1 year after epilepsy clinic attendance and at the time of interview, were compared. Between group comparison was done at the time of interview for seizure control, adverse events (AE), socioeconomic status, earning and marital status to assess the impact of TDM. Comparisons were performed using the unpaired Student's t-test, the chi square test or Fisher's exact test.

Results

Demographic data

There were 16 males and 9 females in each group. Mean ± s.d. age at onset of epilepsy for group A was 21.22 ± 3.26 years and that group B was 21 ± 3.7 years. Mean duration of treatment for group A was 9.68 ± 4.06 years, while that for group B was 8.04 ± 2.58 years. Mean duration of TDM OPD attendance for group A was 4.08 ± 1.34 years and 4.01 ± 1.34 years at the epilepsy clinic while that for group B in the epilepsy clinic 3.99 ± 1.3 years. There was no statistically significant difference between the groups with respect to age at onset of epilepsy, duration of treatment, attendance at the epilepsy clinic at the age at the time of interview (P > 0.05, unpaired Students t-test).

Within group comparison

With respect to seizure control, 23/25 (92%) in group A had uncontrolled epilepsy at the time of TDM referral. At interview, 11/25 had achieved complete seizure control, 10/25 had 50% reduction in seizure frequency while 4/25 (16%) still had uncontrolled epilepsy. The difference in group A pre and post TDM referral was statistically significant (P < 0.05, chi square test). In group B, 1 year after start of treatment, 25/25 (100%) had uncontrolled epilepsy. At interview, 2/25 patients had reached complete control, 11/25 had greater than 50% reduction in seizure frequency, while 12/25 (48%) still had uncontrolled epilepsy. This difference in group B was also statistically significant (P < 0.05, chi square test). With respect to adverse events, in group A, 7/25 (28%) had adverse events prior to TDM referral, which was reduced to 2/25 post TDM (8%) (P = 0.004, Fisher's exact test). In group B, 1 year after onset of treatment, 0/25 (0%) patients had adverse events, while 10/25 (40%) had adverse events at interview (P = 0.004, Fisher's exact test). In group A, 0/25 (100%) patients were earning at the onset of epilepsy, while 19/25 (76%) were earning at the time of interview (P < 0.001, chi square test). In group B, 1 year after drug treatment, 3/25 (12%) were earning, while 12/25 (48%) were earning at the time of interview (P = 0.012). In group A, 0/25 (100%) at the time of TDM referral were married and had children. At interview, 15/25 (60%) were married and had children (P = 0.04, Fisher's exact probability test). In group B, 1 year after onset of treatment, 0/25 (100%) were married with children, while at interview, 7/25 (28%) were married with children (NS).

Between group comparison

Between group comparison at the time of interview showed that there was significant difference in seizure control (Fisher's test P = 0.0032; OR 4.846; 95% CI 1.287,18.255). The incidence of adverse reactions at interview were also significantly different. (P < 0.05, Fisher's exact test).

The cost per patient of the TDM service to the hospital was 147 rupees, while each patient was charged only 30 rupees per sample reflecting the cost of the consumables only (approximately £0.4). The cost of TDM to the hospital per seizure prevented was 22.35 rupees while the cost to the patient was 4.50 rupees (75 Indian rupees = £1).

Discussion

In India, very few laboratories and Clinical Pharmacology centres offer TDM services. In developing countries like ours, with limited resources, the cost of any intervention must be justified in terms of the benefits afforded. Our analysis has shown that although within each group there is a significant benefit by way of improved seizure control, and reduction in adverse events, the group that had undergone TDM showed far greater benefit in terms of reduction in number of seizures (16% vs 48%), number of adverse events (8% vs 40%), better earning status (76% vs 48%) and marital status (60% vs 28%) compared with the group that had not undergone TDM. Since the study was carried out in patients with adult onset epilepsy, earning and marital status were taken as indices of quality of life among the lower socioeconomic class.

The weakness of the study lies in its retrospective design and between group analysis based on a questionnaire and records. In our country, costly diagnostic investigations for epilepsy like the CT scan and magnetic resonance imaging receive priority and TDM may not be requested even once [3]. Thus group B represents a population attending the epilepsy clinic at the same time as group A, but for whichTDM had not been requested.

We use high performance liquid chromatography (h.p.l.c.) technique for analysis of samples. This technique offers significant cost savings in the long term, since it requires a single capital investment and has low recurring costs, which is one of the reasons why the patients are charged only the cost of the consumables. In private hospitals, more expensive techniques such as r.i.a., FPIA and EMIT are used where recurrent costs are fairly high and are charged to patients. Our TDM service also offers a ‘monitoring’ (i.e. interpretation) and not a ‘measuring’ service, while clinical biochemistry laboratories at private hospitals only offer the latter [6].

Our economic analysis carried out from the perspective of a large public hospital, shows that in a chronic disease such as epilepsy, a small additional cost of 60 rupees/year, which is the cost of two TDMs per year, provides significant benefit by way of better seizure control, fewer adverse events and greater chance of remission.