SEARCH

SEARCH BY CITATION

Keywords:

  • Pharmaceuticals;
  • Unused medicines;
  • Landfills;
  • Landfill leachate;
  • POTW discharges;
  • Surface water discharges

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODOLOGY
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES

The pharmaceutical industry is conducting research to evaluate the pathways and fate of active pharmaceutical ingredients from the consumer to surface waters. One potential pathway identified by the researchers is the disposal of unused pharmaceutical products that are discarded by consumers in household trash and disposed of in municipal solid waste landfills. This study was designed to evaluate relative amounts of surface water exposures through the landfill disposal pathway compared to patient use and flushing of unused medicine pathways. The estimated releases to surface water of 24 example active pharmaceutical ingredients (APIs) in landfill leachate were calculated for 3 assumed disposal scenarios: 5%, 10%, and 15% of the total annual quantity of API sold is discarded and unused. The estimated releases from landfills to surface waters, after treatment of the leachate, were compared to the total amount of each example API that would be released to surface waters from publicly owned treatment works, generated by patient use and excretion. This study indicates that the disposal of unused medications in municipal solid waste landfills effectively eliminates the unused medicine contribution of APIs to surface waters; greater than 99.9% of APIs disposed of in a landfill are permanently retained. Integr Environ Assess Manag 2013; 9: 142–154. © 2012 SETAC


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODOLOGY
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES

The detection of trace concentrations of human pharmaceutical compounds in surface water and groundwater continues to receive considerable attention in the technical literature and popular press. The improved precision and accuracy of analytical methods for trace organic chemicals, which includes pharmaceutical products and many other types of consumer products, has led to concerns about potential exposure of humans to these chemicals through the drinking water pathway and to aquatic biota that are in surface waters that receive treated domestic sewage. Research is being conducted by the pharmaceutical industry to evaluate the pathways and fate of active pharmaceutical ingredients from the consumer to surface waters. Potential pathways identified by the researchers are the disposal of unused pharmaceutical products by consumers in household trash that is disposed of in municipal solid waste (MSW) landfills and flushing unused medicines directly to public sewerage systems.

This study compares the relative contributions to the total mass of medicines found in the environment from patient dosing with the mass that results from landfill disposal and from flushing of unused medicines. The definition of unused medicines in this study is limited to unused products that are disposed of by patients or health care providers. Only the active pharmaceutical ingredients (API) in prescription and generic drugs are evaluated in this study. Bulk quantities, surplus, or expired APIs generated by wholesalers or pharmacies are specifically excluded from this analysis because they are returned to their manufacturers and are generally not discarded in municipal landfills. Only landfills defined and regulated as MSW landfills by Subtitle D of the Resources Conservation and Recovery Act (RCRA) are evaluated in this study. Because all active MSW landfill cells in the United States must meet the Subtitle D regulations under federal law, all such landfills are included in this evaluation.

The research compares the API releases in treated landfill leachate to the environmental loadings to surface waters from patient use of pharmaceutical products. In addition, an evaluation of the relative contribution of medicines to surface water based on an assumed scenario where all unused medicines are disposed of by flushing to municipal sewerage systems is evaluated. When the term “conservative” is used to describe an assumption used in this evaluation, it means that the assumption is expected to overestimate the releases of APIs through the pathway being described.

METHODOLOGY

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODOLOGY
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES

Scope

This study evaluated the potential releases of APIs to surface water in landfill leachate, assuming a range of amounts and therapeutic classes of unused pharmaceuticals discarded by consumers. The annual amounts of APIs sold were obtained from IMS Health. The quantities of APIs released in landfill leachate are calculated using a partitioning coefficient and account for biodegradation of the APIs in a landfill. The estimates of potential API releases to surface waters account for the collection and treatment of the landfill leachate at publicly owned treatment works (POTWs) and privately owned treatment plants that comply with the US Environmental Protection Agency (USEPA) effluent limitations guidelines for Subtitle D landfills (40 CFR 445, Subpart B). The estimated releases from POTWs due to patient use and excretion are included for comparison.

Examples of pharmaceutical ingredients

Twenty-three example APIs that represent a range of prescription drugs in terms of sales for human use and physical–chemical properties were chosen for this evaluation because these APIs were included in the 2002 US Geological Survey (USGS) study of APIs and consumer products in US surface waters (Kolpin et al. 2002). One API metabolite of paroxetine hydrochloride was also included in the USGS survey and is included in this evaluation. It is considered to be an API for the purposes of this report, which results in a total of 24 example APIs for this study. Table 1 lists the 24 example APIs evaluated in this study and includes their Chemical Abstracts Service (CAS) numbers, molecular weights, and an example of the mass of each sold annually (based on reports by IMS Health [2001, 2005]). An adjustment factor to convert mass of salt to mass of active ingredient is also shown in Table 1. Salt forms of APIs are often used in formulations because of their stability and other favorable physical properties. However, on ingestion and/or entry into the aquatic environment, the salts will dissociate into their acid or base forms and their behaviors and effects will be a function of those forms. A mass adjustment factor is required to convert the mass of the formulated salt into the mass of the moiety actually present in the environment (only for certain APIs as shown in the table).

Table 1. Annual US sales data for APIs evaluated—2001, 2005
SubstanceCASMWMV adjustment factorMass of API-salt (kg/y)Mass of API purchased (kg/y)
  1. API = active pharmaceutical ingredients; CAS = Chemical Abstracts Service; MV = mass volume; MW = molecular weight.

Acetaminophen103-90-2151.161 5 691 120
Albuterol (salbutamol)18559-94-9239.31 4300 
 Sulfate (2:1)51022-70-9576.70.83 3569
Cimetidine 51481-61-9252.34 57 448 
 Hydrochloride70059-30-2288.810.87 49 980
Ciprofloxacin85721-33-1331.35 94 933 
 Hydrochloride86393-32-0367.810.9 85 440
Codeine 76-57-3299.37 20 127 
 Phosphate52-28-8397.360.75 15 095
 Sulfate6854-40-6397.450.75  
Digoxin 20830-75-5780.951 229
Diltiazem 42399-41-7414.52 162 278 
  Hydrochloride33286-22-5450.980.92 149 296
Doxycycline 564-25-0444.44 38 121 
 Hyclate (HCl, ½ C2H6O, ½ H2O)24390-14-5512.90.86 32 784
Enalapril75847-73-3376.45 1087 
 Maleate76095-16-4492.52   
 Enalaprilat76420-72-9348.40.71 772
Erythromycin114-07-8733.93 65 595 
 H2O 751.940.98 64 283
Fluoxetine 54910-89-3309.33 13 971 
 Hydrochloride56296-78-7345.790.89 12 434
Gemfibrozil 25812-30-0250.341 231 530
Ibuprofen 15687-27-1206.281 1 035 229
Lincomycin 154-21-2406.54 357 
 Hydrochloride859-18-74430.92 328
Metformin 657-24-9129.17 2 048 573 
 Hydrochloride1115-70-4165.630.78 1 597 887
Norfloxacin 70458-96-7319.331 2700
Oxytetracycline 79-57-2460.44 34 
 Hydrochloride2058-46-0496.90.92 31
Paroxetine 61869-08-7329.37 21 400 
 Hydrochloride78246-49-8365.83   
 Metabolite 331.380.91 19 474
Ranitidine66357-35-5314.41 1 115 74 
 Hydrochloride71130-06-8350.870.9 100 417
Sulfamethoxazole723-46-6253.281 314 389
Sulfathiazole72-14-0255.321 483
Tetracycline60-54-8444.44 74 532 
 Hydrochloride64-75-5480.90.92 68 569
Trimethoprim738-70-5290.321 64 450
Warfarin5543-58-8308.33 4300 
 Sodium 331.320.93 3999

Solid and/or liquid partition (adsorption) coefficients are used to predict the extent to which an organic chemical partitions between the solid and solution phases of water and soils, sediments, and other solids including the organic solid wastes in landfills. The water-organic carbon partitioning coefficient (KOC) and the octanol-water partition coefficient (KOW) of a specific organic chemical are both measures of its hydrophobic characteristics and are used as surrogates to estimate the chemical's potential to partition to solids. There have been a number of studies in recent years evaluating sorption of organic constituents in landfills. However, none of these provide partition coefficients for the example APIs in this study.

The ability to estimate the sorption of an API to solids in various media is critical to understanding its environmental fate. Unfortunately, many of the methodologies and relationships used for determining this important parameter, like KOW, were derived from studies with neutral, hydrophobic compounds such as pesticides and industrial chemicals. For these classes of compounds, the primary driver for partitioning behavior of a chemical is its hydrophobicity, or lipophilicity, and most of the relationships explicitly relate the distribution coefficient to the organic C content of the solid.

The assumption is that the partitioning of the chemical will be predominantly onto the organic fraction of the solid. Although this assumption is useful when dealing with neutral, hydrophobic compounds, for large, multifunctional ionic compounds such as many APIs, the partitioning behavior is more complex.

Cunningham (2008) has developed a methodology for calculating partition coefficient (Kp) values for APIs that adsorb to organic solids in wastewater treatment plants (WWTPs). The Kp value used in this evaluation is the ratio of the mass of dissolved API in the landfill leachate to the mass of API adsorbed on the organic solids in the landfill. This method uses the KOW or octanol-water distribution coefficient (DOW) of an API, and its acidic or basic properties, to calculate the Kp value of the compound.

The wastes disposed of in an MSW landfill contain a large fraction of organic solids that will act as an adsorbent in a way that is similar to the organic content of the primary solids and biological solids in a WWTP. Therefore, landfills have a significant potential for adsorption of organic chemicals, including ionic and neutral APIs, and the Kp values derived by Cunningham (2008) are considered an appropriate starting basis for estimating the partitioning of APIs to the solids contained in landfills.

The Kp value for an API is used to calculate its concentration in landfill leachate, based on the assumption that equilibrium occurs between the solid and aqueous phases in a landfill. Given that leachate volumes are low compared to the mass of solids present in a landfill, the equilibrium assumption is realistic.

Sufficient data is available in the technical literature on API chemical properties to calculate Kp values for the 24 example APIs. Table 2 presents the calculated Kp values and related physical and chemical characteristics for each of the 24 APIs evaluated in this study. The greater the magnitude of log Kp, the greater is the propensity of the API to adsorb to solid organic and inorganic materials in a landfill.

Table 2. Partitioning coefficients for APIs evaluated
CompoundDOW or KOWpKaFunctional groupLog Kp

(Cunningham 2008)

  1. API = active pharmaceutical ingredients; DOW = octanol/water distribution coefficient; KOW = octanol/water partition coefficient; na = not available.

Acetaminophen7.769.5Neutral0.003
Albuterol sulfate0.0019.3, 10.3Base0.400
Cimetidine1.586.9Base2.319
Ciprofloxacin0.0036.09, 8.74Zwitterion−1.846
Codeine0.00310.6Base0.686
Digoxin18.1naNeutral0.006
Diltiazem23.18.41Base3.018
Doxycycline0.05375.84, 8.23Neutral0.00002
Enalaprilat0.0791.8, 10.63Zwitterion−1.017
Erythromycin-H2O66.078.8Base3.292
Fluoxetine61.98.7Base3.275
Gemfibrozil1.484.7Acid0.500
Ibuprofen1.074.4Acid0.436
Lincomycin3.37.6Base2.511
Metformin0.05612.4Base1.449
Norfloxacin0.0046.34, 8.75Zwitterion−1.773
Oxytetracycline0.01058.11Base1.013
Paroxetine metabolite21.19.6Base2.995
Ranitidine0.08158.29Base1.547
Sulfamethoxazole7.765.45Base2.734
Sulfathiazole0.3727.2Base1.942
Tetracycline0.0373.3, 8.3, 10.2Base1.342
Trimethoprim4.36.6, 7.12Base2.580
Warfarin29naAcid1.091

Calculation of leachate API concentrations

The methodology assumes that APIs disposed of in landfills are unpackaged and immediately available for dissolution in the liquid phase. This is a conservative assumption (estimates greater concentrations in the leachate than actual) because typically the consumer products will be put in the trash in their packaging. This packaging, which includes plastic bottles and vials, bubble packs, and similar materials, would generally not degrade during the period when the landfill is actively generating leachate (i.e., before it is closed and capped). Therefore, only the amount of API that is present in broken packaging or disposed of without packaging would likely be available for dissolution in the leachate.

The concentration of each API in landfill leachate is calculated using the Kp values in Table 2, which are also included in Table 3. The typical pH in 22 MSWs studied by the USEPA (2002) ranged from 5.9 standard units (SU) to 8.1 SU with an average of 6.9 SU. A pH of 7 SU was used to calculate partitioning to landfill solids in this study. The mass-based ratio of solids-to-liquid in the landfill is 729:1 based on USEPA landfill performance data (USEPA 2002). The leachate concentration is adjusted to account for biodegradation that occurs in the landfill. Biodegradation removal fractions are available for 5 of 24 example APIs: acetaminophen, erythromycin, fluoxetine, ibuprofen, and trimethoprim. Table 3 presents these fractions as well as the values of log Kp for each API. All of the other APIs were assumed to be unaffected by biodegradation in the landfill.

Table 3. Degradation pathways of APIs evaluated
CompoundLog KpLoss by human metabolism (%)Human metabolism referencePrimary and secondary POTW removal (%)POTW removal referenceLandfill biodegradation (fraction remaining)a
  • POTW = publicly owned treatment works.

  • a

    Cunningham et al. 2012.

Acetaminophen0.00310Williams 200598Williams 20050.2000
Albuterol sulfate0.40072PDR 2002–20060No data1.0000
Cimetidine2.31952Williams 200570Williams 20051.0000
Ciprofloxacin−1.84611PDR 2002–200674Lindberg et al. 20061.0000
     Castiglioni et al. 2006 
     Golet et al. 2002 
Codeine0.68610PDR 2002–200646Gómez et al. 20071.0000
Digoxin0.00616PDR 2002–20060No data1.0000
Diltiazem3.01896Williams 200570Williams 20051.0000
Doxycycline0.0000No data0No data1.0000
Enalaprilat−1.01710Williams 200530Williams 20051.0000
Erythromycin-H2O3.2920No data66Karthikeyan and Meyer 20060.5000
Fluoxetine3.27590Williams 200585Williams 20050.9600
Gemfibrozil0.50024Williams 200544Williams 20051.0000
Ibuprofen0.43678Williams 200590Williams 20050.5000
Lincomycin2.51183Rxlist.com 20070Castiglioni et al. 20061.0000
Metformin1.4490No data7Williams 20051.0000
Norfloxacin−1.7737Williams 200581Castiglioni et al. 20061.0000
     Golet et al. 2002 
Oxytetracycline1.0130No data0No data1.0000
Paroxetine metabolite2.9950No data89Williams 20051.0000
Ranitidine1.5476Williams 200530Williams 20051.0000
Sulfamethoxazole2.73488Ruggy 194524Castiglioni et al. 20061.0000
Sulfathiazole1.94215Vree et al. 200480No data1.0000
Tetracycline1.3420No data0Karthikeyan and Meyer 20061.0000
Trimethoprim2.58015Rxlist.com 200729Williams 20050.2000
Warfarin1.09292PDR 2002–20060No data1.0000

Table 3 also includes the fractions of each example API that are metabolized by patients who use the medicines and the removal fractions in wastewater treatment plants. If no metabolism data are available for an API, metabolism is assumed to be zero. These data are used for the estimation of releases of APIs to surface waters after treatment of domestic sewage and landfill leachate.

The wastes disposed of in MSW landfills constitute a heterogenous matrix, in terms of both the size of the solids and the composition of organic and inorganic materials. Landfills are incompletely saturated (i.e., all voids are not filled with water) and thus not all solids in the landfill are in contact with leachate at any specific time. Therefore, the partitioning of the APIs to solids in an MSW landfill may be less efficient than the partitioning that occurs in a biological treatment plant.

To account for this heterogeneity, an additional term for sensitivity analysis of the sorption effect, referred to in this study as “MSW sorption efficiency,” was included in the calculation of the API leachate concentration. A range of MSW sorption efficiencies from 0.01 to 1.0 was examined to evaluate the sensitivity of the predicted leachate API concentrations and masses to the sorption efficiency in the landfill. An MSW sorption efficiency of 1.0 means that the sorptive efficiency in an MSW landfill is equal to that of the solids in a wastewater treatment plant; a sorptive capacity of 0.01 means that the sorptive efficiency in a landfill is 1% of that of wastewater treatment solids.

Leachate volume and API mass releases

The leachate volume is calculated from the quantities presented in a USEPA report on landfill waste containment systems (USEPA 2002). The maximum volume of leachate generated during the active operation periods of landfills in the northeast and southeast United States was used in the calculation. Leachate rates from the northeast and southeast represent the highest leachate rates in the United States and were selected to be conservative. The assumption is made that partitioning is at equilibrium between the aqueous and solid phases, and therefore the greater the volume of leachate the greater the mass of API in the leachate. The maximum leachate volume, which was used for all leachate calculations in this study, is 14 300 liters per day per hectare (L/ha-d) of landfill surface area. As described in the USEPA report, the volume of leachate decreases dramatically once a landfill cell is closed, so use of this maximum leachate volume estimate is highly conservative.

The MSW solids disposal rate is calculated from reported landfill disposal for the year 2010 (USEPA 2011). The disposal rate used in this study is 2.4 lb/d-person (assuming 34.0% average recycle of MSW, as recorded by USEPA). It is also assumed that waste is applied in a 2.5-meter lift per day and at a compacted density of 1200 lb/yd3 is achieved in the landfill (O'Leary and Walsh 1995). These figures represent averages of national data for MSW landfills.

Based on the above assumptions, the rate of leachate generation is calculated as 0.0015 liters/d-capita. For comparison, the average rate of leachate generation for the northeast and southeast US landfills reported by the USEPA (2002) is 0.00035 L/d-capita.

The US population is assumed to be 307 000 000 for the leachate generation calculation. Thus, a total volume of leachate of 168 000 000 L/y is used to estimate the total mass of each API that is leached from MSW landfills. The mass is calculated by multiplying the leachate concentration by the annual leachate volume.

Predicted surface water releases of leachate

Leachate from Subtitle D landfills is treated at on-site wastewater treatment facilities or at (POTWs). Publicly owned treatment works must achieve a minimum of secondary treatment standards as promulgated at 40 CFR 133 and this study assumed that API removals achieved by conventional secondary treatment would be achieved for APIs in leachate. Many POTWs must apply advanced biological treatment (e.g., nitrification, denitrification) or tertiary treatment (e.g., filtration) to achieve water quality-based effluent limits, so the assumption that all APIs in domestic sewage receive secondary treatment is a conservatively low estimate of removal.

Direct discharges to surface water of treated leachate from Subtitle D landfills must achieve the effluent limitations guidelines at 40 CFR 445, Subpart B. These effluent limitations guidelines are based on best practicable control technology (BPT) for conventional pollutants such as biochemical O demand and total suspended solids, and best available technology economically achievable (BAT) for toxic and nonconventional pollutants. BPT is equivalent to secondary treatment for POTWs, and BAT for Subpart B landfills (Subtitle D in RCRA terminology) is the same as BPT. Therefore, it was assumed for this study that privately owned treatment plants for Subtitle D landfill leachate would achieve the same API fractional removals that are achieved by POTWs. This is also a conservative assumption, because treatment systems that are designed and operated to treat landfill leachate would be expected to achieve higher removals of pollutants than a POTW, because the biomass in the landfill leachate treatment system will be acclimated to the specific leachate composition being treated, which may have higher and less variable API concentrations than would a POTW.

To estimate the concentration of each of the example APIs that could enter the environment due to the collection and discharge of leachate, the mass of API discharged after wastewater treatment was calculated using the POTW percent removal data available in the literature. The primary and secondary treatment system removal efficiencies assumed in this study are shown in Table 3.

Contributions by patient use

To put the estimated releases to surface water resulting from the disposal of unused medicines to landfills or disposal by flushing to sewers in context, the releases of APIs due to patient use and subsequent excretion into domestic sewage should be included in the comparison. The methodology used to calculate surface water releases of APIs is described in Anderson et al. (2004). The methodology uses the total annual sales of a specific API and calculates a quantity excreted to sewage by patients based on the fraction of the API that is metabolized. The calculated API loading in the sewage is then adjusted for removal in the primary and secondary treatment processes that are typically used at POTWs (Table 3). The quantity of API released to surface waters in the treated sewage is that which remains after accounting for metabolism by patients and treatment at the POTW. The POTW percent removals used for this calculation are the same as those used to calculate the removals of APIs in landfill leachate that is sent to POTWs or privately owned treatment plants for treatment.

Example calculations

  • 1.
    Concentration of API in landfill leachate, µg/L = (mass of unused API disposed, kg) × (1-fraction biodegraded)(109 µg/kg)/(1.7 × 108 L water + (MSW sorption efficiency) × 2.93 × 1011 L solids × 10log Kp)
  • 2.
    Leachate mass flux, µg/capita-day = (0.0015 L/capita-d leachate) × (concentration of API in leachate, µg/L)
  • 3.
    Mass of API in leachate, kg/y = (leachate mass flux, µg/capita-d) × (3.07 × 108 capita) × (365 d/y)/(109 µg/kg)
  • 4.
    API mass in POTW effluent due to patient use, kg/y = (mass of API purchased kg/y) × (1-fraction of API disposed as unused) × (1-fraction of API metabolized) × (1-fraction of API removed by POTW treatment)
  • 5.
    API mass in POTW effluent due to landfill leachate treatment, kg/y = (mass of API in leachate, kg/y) × (1-fraction of API removed by POTW treatment)

Comparison of calculated leachate concentrations to measured concentrations

The Maine Department of Environmental Protection's (MDEP) study of active pharmaceutical ingredients (API) in the leachate from 3 landfills is a useful addition to the ongoing research effort on the sources and fate of APIs in the environment. The MDEP report, “Preliminary Characterization of the Pharmaceutical Content of Municipal Solid Waste Landfill Leachate from Three Landfills in Maine,” describes the landfills that generated the leachate that was sampled. All 3 landfills had engineered liner and leachate collection systems. This landfill leachate is equivalent to the leachate calculation assumptions in this report although it should be recognized that the nationwide release assumptions developed in this report are not intended to correspond to any individual landfill. The landfill model estimates are intended to serve as an estimate of the total annual releases of APIs for all active Subtitle D landfills.

Thirteen of the APIs measured in the landfill leachate by MDEP were evaluated in this study. Table 4 presents a comparison of the MDEP analyses and the model predictions, assuming a 0.01 sorption efficiency in the landfill, i.e., the highest leachate concentrations that were simulated.

Table 4. Comparison of predicted and measured API leachate concentrations
Parameter nameHatch Hill landfill, AugustaBath municipal landfillBrunswick municipal landfillGeometric mean of measured concentrationsPhRMA predicted at 15% unused leachate concentrationaPhRMA predicted at 10% unused leachate concentrationaPhRMA predicted at 5% unused leachate concentrationa
Conc. (µg/L)Conc. (µg/L)Conc. (µg/L)Conc. (µg/L)Conc. (µg/L)Conc. (µg/L)Conc. (µg/L)
  • API = active pharmaceutical ingredients; Conc. = concentration; PhRMA = Pharmaceutical Research Manufacturers of America.

  • a

    Assumed 0.01 sorption efficiency. Measured values below reporting limit are entered as 50% of the reporting limit.

Acetaminophen117.002.750.032.1318 26336 52654 789
Albuterol0.600.030.090.11244771
Cimetidine0.020.150.060.064.18.212.3
Ciprofloxacin0.270.000.000.0125 47850 95676 435
Diltiazem0.020.000.010.0152105157
Enalapril0.040.000.000.01230460690
Erythromycin-H2O2.990.030.290.300.30.60.8
Gemfibrozil0.170.150.280.19122624533679
Ibuprofen23.2021.9011.6018.06317263449516
Lincomycin0.060.070.280.110.00.00.1
Metformin14.801.180.060.9996819352903
Norfloxacin0.450.020.020.0680516102415
Sulfathiazole0.040.000.260.030.10.20.3

Comparison of the geometric mean of the API concentrations measured by MDEP to the range of concentrations predicted by in this study show that this study's predictions of leachate API concentrations are, with one exception, greater than the measured concentrations from the Maine study, in some cases by multiple orders of magnitude. The geometric mean is used in this comparison because the individual landfill concentrations for a specific API often vary by a factor of 10 or more—use of an arithmetic average in such cases will severely bias the result and is not a reliable estimate of the midpoint of the leachate concentration distributions.

The lincomycin data in Table 1 show annual sales of 357 kg/y, which is the principal reason for the low predicted leachate concentrations by the landfill model (predicted values were 0.02 to 0.05 µg/L). Lincomycin is a veterinary pharmaceutical so it is possible that the Maine landfill concentrations are influenced by veterinary medicines because the 357 kg/y is for sales exclusively for human use.

The MDEP data are based on filtered samples, so as stated in the report, it is possible that total concentrations of APIs measured in the Maine landfill leachate are greater than shown in Table 4 for those APIs that adsorb to solids. However, for many of the APIs measured by MDEP, the concentrations predicted by this study's landfill leachate model are substantially greater than the measured concentrations and the contribution of adsorbed APIs on solids are unlikely to change the measured concentrations enough to invalidate the comparisons in the table.

Groundwater releases

The EPA regulations for Subtitle D landfills (40 CFR 258) establish minimum technology guidelines (MTG) for landfill cells constructed after the effective date of the rule. The MTG for landfill liner systems consists of a permeable leachate collection and removal layer (a minimum 12-inch thickness of granular material), located on top of a composite liner consisting of a low permeability geomembrane on top of a minimum 24-inch thickness of compacted, low-permeability clay. These composite liner systems are designed to be highly effective at preventing leachate migration to groundwater.

A USEPA study (2002) of the performance of Subtitle D landfill liner systems concludes that the required liner systems will substantially prevent leachate migration for the entire period of significant leachate generation for typical landfills. Therefore, for the objectives of this study it was concluded that the landfill-leachate-groundwater release pathway is negligible and no estimates of such releases are practical.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODOLOGY
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES

The estimated releases to surface water of APIs in landfill leachate, for the 24 example APIs, are shown in Table 5 for 1 of the 3 disposal scenarios evaluated in the landfill study. As shown in Table 5, the total annual purchases of these 24 APIs is greater than 9.5 million kg. Metabolism and POTW removal reduce this annual purchased mass by approximately 80% to 82%, for the range of patient use assumed for this study. Table 5 presents the leachate mass loadings to surface water for 3 unused API disposal rates—5%, 10%, and 15% of total annual quantity sold is assumed to be discarded unused to landfills—and compares the landfill leachate loadings in treated effluents to the total amount of each API released to surface waters after patient use and wastewater treatment. The assumed MSW sorption efficiency is 0.01, which results in the greatest estimated releases of APIs in the leachate.

Table 5. Impact of landfill disposal on total surface water load of selected pharmaceutical active ingredients due to patient dosing (0.01 sorption efficiency)
CompoundAPI quantity purchased (kg/y)Loss by human metabolism (%)aPOTW removal (%)a5% API disposed10% API disposed15% API Disposed
API mass in POTW effluent due to patient use (kg/y)API mass in POTW effluent from unused medicine in landfills (kg/y)Total load resulting from landfill disposal (%)API mass in POTW effluent due to patient use (kg/y)API mass in POTW effluent from unused medicine in landfills (kg/y)Total load resulting from landfill disposal (%)API mass in POTW effluent due to patient use (kg/y)API mass in POTW effluent from unused medicine in landfills (kg/y)Total load resulting from landfill disposal (%)
  • API = active pharmaceutical ingredients; POTW = publicly owned treatment works.

  • a

    See references provided in Table 3.

Acetaminophen5 691 120109897 31861.20.0692 196122.50.1387 074183.70.21
Albuterol sulfate35697209494.00.428997.90.8884911.91.38
Cimetidine49 980527068370.20.0064770.40.0161180.60.01
Ciprofloxacin85 440117418 7821110.75.5817 7942221.411.1016 8053332.116.55
Codeine15 095104669694.70.0766039.50.14623614.20.23
Digoxin2291601830.60.331731.20.701641.81.11
Diltiazem14 9296967017020.10.0116120.20.0215230.40.02
Doxycycline32 7840031 14588.70.2829 506177.40.6027 866266.10.95
Enalaprilat772103046227.05.5243854.010.9941381.016.39
Erythromycin-H2O64 28306620 7630.00.0019 6710.00.0018 5780.00.00
Fluoxetine12 43490851770.00.001680.00.001590.00.01
Gemfibrozil231 530244493 612115.20.1288 685230.30.2683 758345.50.41
Ibuprofen1 035 229789021 63653.20.2520 498106.40.5219 359159.60.82
Lincomycin328830530.00.01500.00.01470.00.02
Metformin1 597 887071 411 733150.90.011 337 431301.70.021 263 130452.60.04
Norfloxacin270078145325.75.3642951.310.6740677.015.95
Oxytetracycline3100300.00.03280.00.06270.00.10
Paroxetine metabolite19 47408920350.00.0019280.00.0018210.00.00
Ranitidine100 41763062 7705.70.0159 46711.40.0256 16317.10.03
Sulfamethoxazole314 389882427 2391.30.0025 8052.50.0124 3713.80.02
Sulfathiazole4831580780.00.00740.00.01700.00.01
Tetracycline68 5690065 1418.90.0161 71217.80.0358 28426.70.05
Trimethoprim64 450152936 9510.10.0035 0060.10.0033 0610.20.00
Warfarin39999203040.90.302881.80.642722.81.01
Aggregate9 544 488  1 907 32416590.091 806 93833180.181 706 55349770.29

The APIs shown in Table 5 with the highest potential mass releases to surface water through the landfill leachate pathway are those with low partitioning coefficients (log Kp <1.0), no biodegradation in landfills, and minimal or no removal at a POTW.

The estimated leachate release to surface water of acetaminophen, which has the largest sales of any of the 24 example APIs, is very low because although this API has a low log Kp, its mass in leachate is predicted to be reduced substantially by biodegradation in an MSW landfill and the quantity remaining in the leachate is very effectively biodegraded in the leachate treatment step.

Ibuprofen has relatively high POTW removal (90%), but is sold in very large quantities, has a moderately low Kp, and has data for biodegradation showing 50% reduction in landfills. It is significantly metabolized by patients (70%), and therefore the proportion of ibuprofen in the discharge to surface water that originates from unused medicine disposal in landfills is higher than that of acetaminophen.

Table 5 also compares the annual API mass released to surface water through landfill disposal to the total mass of API released to surface water from patient use. Even at the greatest assumed disposal rate of unused medicines in landfills, the landfill leachate pathway to surface water is dwarfed by the surface water releases due to patient use and excretion of the 24 example APIs.

The sensitivity analysis of the leachate API mass discharged to surface water (after metabolism and POTW treatment for the quantity used by patients) is carried out using the MSW sorption efficiency and the quantity of unused medicines disposed of in landfills and is shown graphically in Figure 1. The purpose of Figure 1 is to demonstrate that the calculated mass releases of API in landfill leachate are not very sensitive to changes in either the mass of medicine discarded unused or the efficiency of the sorption process in landfills. A 100-fold decrease in MSW sorption efficiency results in about a 8-fold increase in the leachate API mass discharged to surface water. This analysis indicates that the sensitivity of the leachate API mass to the sorption efficiency of the materials in the landfill is much less than 1:1. As expected, the total mass of API in the leachate is a linear function of the quantity of API disposed of with municipal trash. The calculated total annual releases of API to surface water ranges from 0.06% to 0.29% of the releases due to patient use of the medicines.

thumbnail image

Figure 1. Sensitivity of landfill mass releases to quantity disposed and sorption efficiency.

Download figure to PowerPoint

Disposal of unused medicines by flushing them to public sewerage systems has been a historically recommended practice. Comparison of the relative contribution of APIs to surface water from landfill disposal of unused medicines to disposal of medicines by flushing to the sewer demonstrates the relative efficiency of the 2 approaches for reducing API contributions of unused medicine to surface water.

For this comparison, the estimated aggregate annual surface water releases of the 24 example APIs were calculated by assuming that all unused medicines were disposed of in public sewerage systems. The total surface water release in this calculation is a result of patient use and excretion and disposal of unused medicines to the sewerage systems. The surface water discharge estimates are calculated using the methodology described for landfills (POTW treatment). All API mass quantities are adjusted, as appropriate, to account for the salt fraction of the product if the product is distributed as a salt. The discharges from POTWs due to patient use are based on the mass of API sold, the percentage metabolized, and the percent removed by POTW treatment.

Table 6 presents the calculated surface water discharges of the 24 example APIs assuming unused medicine disposal rates of 5%, 10%, and 15% of annual purchases and a 0.01 sorption efficiency. The percent increase in the total annual surface water discharge of the 24 example APIs that would be caused by unused medicine disposal in sewers, using landfill disposal of unused medicines as the base case for comparison, is 12.2%, 20.3%, and 27.0% for the 5%, 10%, and 15% disposal rates, respectively. This analysis indicates that encouraging the disposal of unused medications in municipal solid waste landfills will decrease the surface discharges of APIs compared to flushing unused medicines to sewers.

Table 6. Impact of unused medicine disposal method on total surface water load of selected pharmaceutical active ingredients
Compound5% API disposed10% API disposed15% API disposed
Unused API discarded to sewer (kg/y)API mass in POTW-treated effluent (kg/y)a,bAPI mass from used medicine disposal (kg/y)Total load resulting from sewer disposal (%)Unused API discarded to sewer (kg/y)API mass in POTW-treated effluent (kg/y)a,bAPI mass from used medicine disposal (kg/y)Total load resulting from sewer disposal (%)Unused API discarded to sewer (kg/y)API mass in POTW-treated effluent (kg/y)aAPI mass from used medicine disposal (kg/y)Total load resulting from sewer disposal (%)
  • API = active pharmaceutical ingredients; POTW = publicly owned treatment works.

  • a

    Includes contribution from patient use and excretion.

  • b

    See references provided in Table 3.

Acetaminophen 284 556103 00956915.5569 112103 57811 38211.0853 668104 14717 07316.4
Albuterol sulfate178112817815.8357125635728.4535138553538.7
Cimetidine 249975877509.949987977149918.874978367224926.9
Ciprofloxacin427276 51142725.6854476 981854411.112 81677 45112 81616.5
Codeine 75513 6617555.5151013 737151011.0226413 812226416.4
Digoxin11194115.9231962311.7341983417.4
Diltiazem 74653941223956.814 9306091447973.522 3948241671881.5
Doxycycline163932 78416395.0327832 784327810.0491832 784491815.0
Enalaprilat 39489275.5774925411.01164948116.4
Erythromycin-H2O321464 28332145.0642864 283642810.0964264 283964215.0
Fluoxetine6222709334.5124335418752.6186543828063.8
Gemfibrozil11 577100 09564836.523 153101 65112 96612.834 730103 20719 44918.8
Ibuprofen 51 76126 812517619.3103 52330 85010 35233.6155 28434 88715 52844.5
Lincomycin16328165.0333283310.0493284915.0
Metformin 79 8941 486 03574 3025.0159 7891 486 035148 60310.0239 6831 486 035222 90515.0
Norfloxacin13525201355.4270253027010.7405253940515.9
Oxytetracycline23125.0331310.0531515.0
Paroxetine metabolite 97421421075.01947214221410.02921214232115.0
Ranitidine 502166 28535155.310 04266 496702910.615 06266 70710 54415.8
Sulfamethoxazole15 71951 56015 71930.531 43965 39331 43948.147 15879 22647 15859.5
Sulfathiazole24414245.8484184811.6724217217.2
Tetracycline342868 56934285.0685768 569685710.010 28568 56910 28515.0
Trimethoprim322339 23922885.8644539 582457611.6966839 925686417.2
Warfarin 20050420039.740068840058.160087260068.8
Aggregate477 2242 148 394130 2666.1954 4492 172 443260 53212.01 431 6732 196 492390 79817.8

Figure 2 compares the surface water discharges of the 24 APIs that were evaluated for 3 cases: 1) all unused medicine disposal is by flushing to the sewer, 2) all unused medicine disposal is to municipal solid waste landfills, and 3) unused medicine is disposed of elsewhere (i.e., not flushed to the sewer or sent to a landfill) and surface water releases are solely due to patient use.

thumbnail image

Figure 2. Comparison of surface water discharges from landfill and sewer disposal of unused medicines (0.5 sorption efficiency).

Download figure to PowerPoint

These are the boundary cases for unused medicine disposal (i.e., 100% of unused medicines going to one or the other disposal route, not split between both). The quantities of unused medicines are based on 3 percentages of total annual sales (in kg) for each of the 24 APIs evaluated. The contribution of patient use of the medicines, with unused medicines disposed of elsewhere, is also shown on Figure 2 for comparison. As shown, if unused medicines are disposed of in landfills the total surface water contribution is essentially the same as the contribution from patient use alone.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODOLOGY
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES

The 2 most convenient methods that patients can use to dispose of unused medicines are: 1) disposing of them with household trash, or 2) flushing them down the toilet. In some locations, take-back programs are beginning to be available to the public. However, participation in these programs is low. The relative importance of the 2 most common disposal methods for unused medicines to total surface water discharges of the 24 example APIs is shown in Table 6. The comparison of the calculated mass discharges is presented in Table 7.

Table 7. Relative importance of sewer flushing and landfill disposal of unused medicines (0.5 MSW sorption efficiency)
Unused medicine disposal methodPercent of total surface water release due to specified disposal method of unused medicinesa,b
5% unused10% unused15% unused
  • API = active pharmaceutical ingredients; MSW = municipal solid waste; POTW = publicly owned treatment works.

  • a

    Includes API contribution from patient use and excretion.

  • b

    See references provided in Table 3.

Sewer flushing6.112.017.8
Landfill0.090.180.29

The landfill contribution to surface water from unused medicines is calculated by accounting for partitioning of each API to the organic and inorganic solids in the landfill and biodegradation of the API in the landfill, if applicable. It is assumed that the leachate is transferred to a POTW for treatment or treated and discharged at the landfill site. The surface water contribution of unused APIs due to flushing to the sewer is calculated by assuming that the unused mass of API is discharged to the sewer without any metabolism of the chemical. The unused API contribution to the surface discharge is calculated as the mass of unused medicine at each percentage disposal rate multiplied by its percent removal at the POTW.

As shown in Table 7, if unused medicines are flushed to the public sewerage systems such disposal would constitute 6.1%, 12.0%, and 17.8% of the total surface water discharges (including patient use and excretion) for the 5%, 10%, and 15% unused medicine quantities, respectively. The reason why the unused medicine disposal in the sewer causes a larger increment of surface water releases than the unused percentage of total API purchases is that patient use includes metabolism of a number of the APIs before they are excreted, thus reducing the total quantities of those APIs that are sent to the sewer.

Patient use of medicines is the principal source of the surface water discharges of APIs regardless of the disposal method for unused medicines. Landfill attenuation of APIs and subsequent landfill leachate treatment by POTWs or BAT facilities results in substantially lower estimated total discharges of API to surface waters when unused medicines are disposed of by landfilling as opposed to by flushing to the public sewers. Essentially, there is no difference in the surface water releases of these 24 APIs between the disposal of unused medicines in landfills and disposal of unused medicines elsewhere.

CONCLUSIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODOLOGY
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES

The objective of this study was to estimate the potential releases of APIs to surface water resulting from the disposal of unused medicines in Subtitle D (MSW) landfills. Based on the evaluation carried out for 24 APIs, it can be concluded that:

  • 1.
    This study indicates that the disposal of unused medications in municipal solid waste landfills effectively eliminates the unused medicine contribution of APIs to surface waters.
  • 2.
    The landfill disposal pathway to surface water accounts for an average of 0.29% to 0.06% of the estimated aggregate annual surface water releases for the 24 APIs evaluated by this study (at an assumed landfill sorption efficiencies of 0.01 and 1.0, respectively and for unused medicine accounting for 5% to 15% of total medicine sales). These landfill mass contributions are based on conservative assumptions of landfill leachate generation that would tend to predict higher leachate concentrations.
  • 3.
    If all unused medicines (5% to 15% of total sales) are disposed of in landfills, a very conservative estimate of 99.7% to 99.94% of API surface water releases would be due to patient excretion.
  • 4.
    If all unused medicines were disposed of by flushing to the sewer, then unused medicine disposal would constitute approximately 6.1%, 12.0%, and 17.8% of the total surface water discharges of these 24 APIs at 5%, 10%, and 15% rates of unused medicine disposal (as a function of annual sales), respectively.
  • 5.
    Notwithstanding the fact that the solids in a landfill are more heterogeneous than those in a wastewater treatment plant, the sensitivity analysis carried out using the sorption efficiency factor (that was varied by a factor of 100) that was used in this study to evaluate this effect demonstrated that the total annual mass releases of API in landfill leachate are relatively insensitive to the partitioning coefficient magnitude. The 100-fold variation in assumed partitioning caused only an 8-fold variation in calculated mass releases of total APIs in the landfill leachate.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODOLOGY
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES

The Pharmaceutical Research Manufacturers of America (PhRMA) funded this study and many member company representatives participated in the review of the methodology and the study results. This study extends work on landfill disposal of APIs that was performed for PhRMA by Golder Associates (2004).

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODOLOGY
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES
  • Anderson PD, D'Aco VJ, Shanahan P, Chapra SC, Buzby ME, Cunningham VL, DuPlessie BM, Hayes EP, Mastrocco FJ, Parke NJ. 2004. Screening analysis of human pharmaceutical compounds in U.S. surface waters. Environ Sci Technol 38:838849.
  • Behr R, Stahler D, Pistell A. 2009. Preliminary characterization of the pharmaceutical content of municipal solid waste landfill leachate from three landfills in Maine. Augusta (ME): Maine Department of Environmental Protection.
  • Castiglioni S, Bagnati R, Fanelli R, Pomati F, Calamari D, Zuccato E. 2006. Removal of pharmaceuticals in sewage treatment plants in Italy. Environ Sci Technol 40:357363.
  • Cunningham VL. 2008. Special characteristics of pharmaceuticals related to environmental fate. In: Kuemmer K, editor. Pharmaceuticals in the environment. 3rd ed. Hidelberg: Springer. p 1324.
  • Cunningham VL, D'Aco VJ, Pfeiffer D, Anderson PD, Buzby ME, Hannah RE, Jahnke J, Parke NJ. 2012. Predicting concentrations of trace organic compounds in municipal wastewater treatment plant sludge and biosolids using the PhATE model. Integr Environ Assess Manag 8:530542.
  • Golder Associates. June 2004. Final report disposal of unused medicines in Subtitle D landfills. Project No: 043-6246. Cherry Hill (NJ).
  • Golet EM, Alder AC, Giger W. 2002. Environmental exposure and risk assessment of fluoroquinolone antibacterial agents in wastewater and river water of the Glatt Valley Watershed, Switzerland. Environ Sci Technol 36:36453651.
  • Gómez MJ, Martínez Bueno MJ, Lacorte S, Fernández-Alba AR, Agüera A. 2007. Pilot survey monitoring pharmaceuticals and related compounds in a sewage treatment plant located on the Mediterranean coast. Chemosphere 66:9931002.
  • IMS Health, Inc. 2001. Reports. IMS Midas Quantum July 2000–August 2001. Collegeville (PA): IMS Health.
  • IMS Health, Inc. 2005. Reports. IMS Midas Quantum July 2004–August 2005. Collegeville (PA): IMS Health.
  • Karthikeyan KG, Meyer MT. 2006. Occurrence of antibiotics in wastewater treatment facilities in Wisconsin, USA. Sci Total Environ 361:196207.
  • Kolpin DW, Furlong ET, Meyer MT, Zaugg SD, Barber LB, Buxton HT. 2002. Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams, 1999–2000 : A national reconnaissance. Environ Sci Technol 36:12021211.
  • Lindberg RH, Olofsson U, Rendahl P, Johansson MI, Tysklind M, Andersson BAV. 2006. Behavior of fluoroquinolones and trimethoprim during mechanical, chemical, and active sludge treatment of sewage water and digestion of sludge. Environ Sci Technol 40:10421048.
  • O'Leary P, Walsh PW. 1995. Decision maker's guide to solid waste management Vol II. Washington (DC): EPA, Office of Solid Waste.
  • [PDR] Physicians Desk Reference. 2002–2006. Electronic version. Greenswood Village (CO): Thomson Micromedex.
  • Ruggy GH. 1945. A review of problems of sulfonamide chemotherapy. Ohio J Sci 45:115124.
  • Rxlist.com. 2010. Entry for Trimethoprim. [cited 2011 September.] Available from: http://www.rxlist.com/trimethoprim-drug.htm
  • Tischler L. 2007. Potential contribution of unused medicines to environmental concentrations of pharmaceuticals. Round Rock, TX: Pharmaceutical Research and Manufacturers of America.
  • [USEPA] US Environmental Protection Agency. 2002. Assessment and recommendations for improving the performance of waste containment systems. Cincinnati (OH): USEPA. EPA/600/R-02/099.
  • [USEPA] US Environmental Protection Agency. 2011. Municipal solid waste generation, recycling, and disposal in the United States: Facts and figures for 2010. [cited 2011 November.] Available from: http://www.epa.gov/osw/nonhaz/municipal/pubs/msw_2010_factsheet.pdf
  • Vree TB, van der Ven AJAM, Verwey-van Eissen CPWGM, van Ewijik-Beneken EWJ, Kolmer EWJ, Swolfs AEM, van Galen PM, Amatdjais-Groenen H. 2004. Isolation, identification and determination of sulfamethoxazole and its known metabolites in human plasma and urine by high-performance liquid chromatography. J Chromatogr B 658:327340.
  • Williams RT. 2005. Human pharmaceuticals: Assessing the impacts on aquatic ecosystems. Pensacola (FL): SETAC. p 71110.