Pathogen inactivation and quantitative microbial risk assessment for Peepoo sanitation system, Kibera

UPTEC W 20039

Examensarbete 30 hp September 2020

Pathogen inactivation and quantitative microbial risk assessment for

Peepoo sanitation system, Kibera

Linnea Eriksson

Lisa Sundberg

ABSTRACT

Pathogen inactivation and quantitative microbial risk assessment for Peepoo sanita- tion system, Kibera

Linnea Eriksson, Lisa Sundberg

Unsafe sanitation systems poses a risk for pathogen transmission, wherefore it is impor- tant to both inactivate pathogens present in human excreta and conduct safe sanitation systems from use to end-use. The Peepoo toilet, using ammonia sanitisation, have been suggested as a low-cost sanitation solution and is implemented in schools in Kibera, an urban slum in Kenya. This master thesis aim to study the inactivation efficiency of am- monia sanitisation when treating human excreta with urea, and to quantify the risks of exposure to microbial hazards from the Peepoo sanitation system using faecal indicator bacteria.

Excreta was collected from four schools in Kibera. After adding urea to mimic the inacti- vation of the Peepoo in the laboratory, the inactivation rate was correlated to temperature and free ammonia concentration for Campylobacter spp., Escherichia coli and Entero- coccus spp.. Campylobacter spp. and E. coli both had a high inactivation rate even at low temperature and low addition of urea. Inactivation rate of Enterococcus spp. was slower and close to zero when 1.87 % urea was added for 15 ^◦ C. For Enterococcus spp.

a lag-phase was observed, which was not affected by temperature but by concentration of free ammonia. For investigated bacteria, inactivation rate increased with increased tem- perature and free ammonia concentration.

Along the Peepoo management chain, several hazardous events were identified such as contamination during usage, handling and transportation. Sampling showed a higher con- tamination of Enterococcus spp. than of E. coli. Enterococcus spp. was used as a faecal indicator for Ascaris and E. coli was used as an indicator of E. coli O157:H7 in a quan- titative microbial risk assessment (QMRA). Through the QMRA, the risk of infection of Ascaris and E. coli O157:H7 for one exposure event was simulated based on a Exponen- tial and a Beta-Poisson dose-response model respectively. The risk of infection of Ascaris was around 12 % regardless of where exposure occurs, if Ascaris eggs were present. For risk for infection with E. coli O157, the simulated risks were below 10 % at almost all exposure points, with most of the high risk exposure points located in the schools.

There are risks of pathogen transmission in the Peepoo management chain that should be further investigated. Ammonia sanitisation permits a high degree of microbial inacti- vation but to secure a safe end-product it is recommended to be kept in room temperature (24.05±0.62 ^◦ C) or higher.

Keywords: Quantitative microbial risk assessment, ammonia sanitisation, faeces, risk assessment, inactivation model, pathogen, Ascaris, E. coli, Campylobacter, Enterococ- cus, sanitation system, sanitisation technology

Department of Energy and Technology, SLU

Box 7032, SE 75007 Uppsala. ISSN 1401-5765

REFERAT

Inaktiveringskonstanter f¨or fekala bakterier och kvantitativ mikrobiell riskanalys f¨or Peepoo:s sanit¨ara system i Kibera

Linnea Eriksson, Lisa Sundberg

Bristande sanit¨ara system utg¨or en risk f¨or spridning av patogener, varf¨or det ¨ar viktigt b˚ade att avd¨oda fekala patogener och att uppr¨atta system som ¨ar s¨akra fr˚an anv¨andning till slutprodukt. Peepoo ¨ar en sanit¨ar l¨osning som anv¨ander sig av ammoniakhygienisering och har implementerats i skolor i Kibera, en urban slum i Kenya. Denna masteruppsats syftar till att studera inaktiveringskonstanten f¨or ammoniakhygienisering vid behandling av m¨anskliga fekalier med urea, samt att kvantifiera risken f¨or exponering av mikrobiella patogener fr˚an Peepoos sanit¨ara system genom att anv¨anda fekala indikatorbakterier.

Inaktiveringskonstanten korrelerades med temperatur och ammoniakkoncentration f¨or Campy- lobacter spp., E. coli och Enterokocker spp.. Avf¨oringen samlades in fr˚an fyra skolor i Kibera, varav de mest infekterade proverna blandades och analyserades p˚a laboratoriet.

Efter att ha adderat urea till avf¨oringen f¨or att efterlikna inaktiveringen som sker i Peep- oon visade det sig att Campylobacter spp. och E. coli b˚ada hade en h¨og inaktivering- shastighet ¨aven vid l˚ag temperatur och l˚ag koncentration av tillsatt urea. Inaktiveringen f¨or Enterokocker var l˚angsammare och n¨ara 0 CFU/dag d˚a 1.87 % urea tillsattes i 15

◦ C. F¨or Enterokocker observerades en laggfas som p˚averkades av koncentration av am- moniak men inte av temperatur. F¨or unders¨okta bakterier ¨okade inaktiveringskonstanten med ¨okad temperatur och koncentration ammoniak.

Flera risker f¨or mikrobiell spridning identifierades l¨angs Peepoos hanteringskedja. Prov- tagning vid anv¨andningsomr˚ade och transport visade p˚a h¨ogre f¨ororening av Enterokocker

¨an av E. coli. Enterokocker anv¨andes som en fekal indikator for Ascaris och E. coli anv¨andes som en indikator f¨or E. coli O157:H7 i en kvantifierad mikrobiell riskanaly (QMRA).

Genom en QMRA simulerades risken f¨or att bli infekterad av Ascaris och E. coli 0157 (EHEC) vid ett exponeringstillf¨alle. Infektionsrisken f¨or Ascaris ber¨aknades till ungef¨ar 12 % oberoende typ av exponering. F¨or E. coli 0157 var den simulerade infektionsrisken l¨agre ¨an 10 % f¨or alla exponeringspunkter, d¨ar de h¨ogsta riskerna fanns inuti skolorna.

Risker f¨or spridning av fekala patogener i Peepoos hanteringskedja b¨or unders¨okas yt- terligare. Ammoniakhygienisering visar p˚a en h¨og avd¨odningsgrad men f¨or att uppn˚a en s¨aker restprodukt b¨or hygienisering ske vid rumstemperatur (24.05±0.62 ^◦ C i f¨orelig- gande studie) eller h¨ogre.

Nyckelord: Kvantifierad mikrobiell riskanalys, ammoniakhygienisering, fekalier, risk- analys, inaktiveringsmodellering, patogen, Ascaris, E. coli, Campylobakter, Enterokocker, sanitetssystem, hygieniseringsteknik

Institutionen f¨or energi och teknik, SLU

Box 7032 SE 75007 Uppsala. ISSN 1401-5765

PREFACE

This final thesis is part of the MSc degree in Environmental and Water Engineering at Up- psala University and Swedish University of Agricultural Sciences. The study corresponds to 30 Swedish academic credits and was conducted as a Minor Field Study, funded by the Swedish International Development Agency (SIDA). Doctor Annika Nordin at Swedish University of Agricultural Sciences and Doctor Nduhiu Gitahi at University of Nairobi were supervisors, and Professor Bj¨orn Vinner˚as at Swedish University of Agricultural Sciences was subject reader.

First, we would give our thanks to both our supervisors for the support and guidance through the whole project. Special thanks to Dr. Annika Nordin, for following us to Kenya and helping us starting up the project, and always was ready to answer our ques- tions no matter day or time. And of course, we will always be grateful to Dr. Nduhiu Gitahi, for making our time in Nairobi so easy, and for making this study possible. Thank to both of you for all the discussions and inputs along the way.

We also want to take the opportunity to thank Mr. Macharia for helping us out in the lab and for being with us at University of Nairobi during the weekends. We wouldn’t have been able do conduct this study without you. Also, we want to thank all members of the lab team, that through the whole time in Nairobi assisted and helped us. Thank for your hospitality and for patiently sharing your knowledge with us.

Last, we want to thank each other for completing this work together. It have been a true experience and a lot of fun!

Linnea Eriksson and Lisa Sundberg Uppsala, October, 2019

Copyright c Linnea Eriksson, Lisa Sundberg and Department of Energy and Technology;

Environmental Engineering Unit, SLU.

UPTEC W 20039, ISSN 1401-5765.

Published digitally at DiVA, 2020, by the Department of Earth Sciences, Uppsala Univer-

sity. http://www.diva-portal.org/

POPUL ¨ ARVETENSKAPLIG SAMMANFATTNING Peepoo - en l¨osning p˚a toalettbristen i Afrikas slum?

Linnea Eriksson, Lisa Sundberg

Idag lever en av tre m¨anniskor v¨arlden ¨over utan tillg˚ang till grundl¨aggande sanitet. Detta inneb¨ar att dessa m¨anniskor dagligen uts¨atts f¨or risk att drabbas utav sjukdomar, s¨arskilt drabbade ¨ar kvinnor och barn. Barn riskerar att insjukna i diarr´e, vilket ˚arligen d¨odar ¨over en halv miljon barn under fem ˚ar, och unga kvinnor kan tvingas att sluta g˚a i skolan n¨ar de f˚ar sin mens. I dagsl¨aget ¨ar vi l˚angt ifr˚an att n˚a FN:s globala m˚al att rent vatten och sanitet ska vara tillg¨anglig f¨or alla till ˚ar 2030. I Afrikas mest t¨atbefolkade slum, Kibera, ¨ar prob- lemen allvarliga: D¨ar har ett av tre barn diarr´e och kvinnor riskerar att bli v˚aldtagna om de vill anv¨anda de publika dassen nattetid. Under regnperioden sv¨ammar dassen ¨over och parasiter och sjukdomsalstrande bakterier sprids med avf¨oringen ¨over de tr˚anga jordtram- pade gatorna. I Kibera finns ingen plats f¨or att bygga toaletter i varje hem. Inte heller kan t¨omningsbilar ta sig in i stora delar av slummen d˚a v¨agarna inte ¨ar tilltr¨ackligt breda. Det kr¨avs en annan l¨osning f¨or att minska sjukdomsspridningen och f¨or att minska risken f¨or att bli utsatt f¨or v˚aldsbrott. En s˚adan l¨osning ¨ar Peepoo-p˚asen. Det ¨ar en nedbrytbar p˚ase som s¨atts p˚a en potta, varp˚a den knyts ihop efter anv¨andning. I p˚asen finns ett substrat, urea, som bryts ned n¨ar det kommer i kontakt med avf¨oringen och renar den fr˚an farliga mikroorganismer. Produkten anv¨ands redan idag p˚a ett antal skolor i Kibera. M˚alet ¨ar att kunna ˚ateranv¨anda avf¨oringen som g¨odsel vid odling, vilket g˚ar hand i hand med FN:s andra globala m˚al: ingen hunger. F¨or att s¨akerhetsst¨alla att avf¨oringen renas tillr¨ackligt, och Peepoo-p˚asen inte ska vara en k¨alla till smittspridning f¨or anv¨andarna, s˚a har tekniken testats p˚a n˚agra vanliga bakterier och en riskanalys har genomf¨orts.

Det visade sig att tekniken som anv¨ands i Peepoo d¨odade tv˚a av tre av de studerade bak- terierna redan inom loppet av tio dagar. F¨or Enterockocker var behandlingen d¨aremot inte lika effektiv. Det visar p˚a att Peepoo-p˚asarna b¨or inneh˚alla h¨ogre koncentrationer av urea

¨an vad som anv¨andes i den h¨ar studien f¨or att avd¨oda samtliga bakterier inom tv˚a m˚anader.

Dock finns det en risk att parasiter ¨overlever l¨angre ¨an s˚a, varf¨or det rekommenderas yt- terligare studier innan slutprodukten kan anv¨andas som g¨odsel.

Peepoo-p˚asarna h¨amtas fr˚an skolorna av tv˚a arbetare, tv˚a g˚anger i veckan. D¨ar jeepen inte kommer in f˚ar de dra stora p˚asar fyllda med Peepo p˚a en rullvagn. Arbetarna anv¨ander handskar och skyddskl¨ader, d˚a det finns risk f¨or att det l¨acker v¨atska fr˚an de insamlade p˚asarna. Skolorna sk¨oter sj¨alva hanteringen av p˚asarna fram tills att de samlas upp av arbetarna, och p˚a m˚anga skolor knyter barnen sj¨alva p˚asen efter anv¨andning. Skolorna f˚ar tv˚al av Peepoople f¨or att bibeh˚alla goda rutiner kring handhygien. Riskanalysen visade p˚a att det finns risk f¨or att bli smittad av patogener vid vistelse i skolorna.

Overlag kan Peepoo ses som en l¨osning f¨or att s¨akerhetsst¨alla sanitet i slumomr˚aden. ¨ S¨arskilt l¨amplig ¨ar sanitetsl¨osningen p˚a skolor, d¨ar utrymmet ¨ar knappt och Peepoo- pottorna m¨ojligg¨or f¨or skolorna att tillgodose tillr¨ackligt m˚anga toaletter f¨or sina elever.

F¨or att bibeh˚alla sanitet st¨alls dock krav p˚a handhygien och p˚a hur p˚asarna hanteras och

f¨orvaras. Studien visar att en behandling under 2 m˚anader i 25 ^◦ C ¨ar tillr¨acklig f¨or att

inaktivera de mer resistenta bakterierna.

Eriksson and Sundberg both planned and performed the study and revised the report. Ob- jectives and conclusions were written by both authors. Main responsibility of this master thesis was as follows:

Lisa Sundberg: Background: Disease transmission from human excreta, Pathogens in human excreta, Risk assessment Methodology: Risk assessment Results: Risk assess- ment Discussion: Risk assessment

Linnea Eriksson: Background: Sanitation systems in Kibera, Onsite treatment of hu- man faeces, Chemical disinfection with ammonia Methodology: Urea treatment Results:

Inactivation rates Discussion: Inactivation rates

ABBREVIATIONS

Word Description

BPW Buffered Peptone Water

CFU Colony Forming Unit

D-value Time to achieve a 90 % inactivation of microorganisms. Also called t ₉₀ . DALY Disability-adjusted life years

EHEC Enterohaemorrhagic Escherichia coli EPEC Enteropathogenic Escherichia coli ETEC Enterotoxigenic Escherichia coli FIB Faecal indicator bacteria

k Inactivation constant

log 10 reduction The relationship between initial and end pathogen concentration, in terms of log ₁₀ N _t - log ₁₀ N ₀

N _t Pathogen concentration at time t N 0 Initial pathogen concentration SSP Sanitation Safety Planning

TAN Total Ammonia Nitrogen

t 90 Decimal reduction time (same as the D-value)

t ₉₉ Time to achieve a 99 % inactivation of microorganisms

TS Total solids

QMRA Quantitative Microbial Risk Assessment

VIP Ventilated Improved Pit latrine

CONTENTS

Abstract . . . . I Referat . . . . II Preface . . . . III Popul¨arvetenskaplig sammanfattning . . . . IV Abbreviations . . . . VI

1 Introduction 1

1.1 Problem description . . . . 2

1.2 Objective . . . . 3

2 Background 3 2.1 Disease transmission from human excreta . . . . 3

2.2 Studied enteric Pathogens . . . . 5

2.2.1 Salmonella spp. . . . . 5

2.2.2 E. coli . . . . 6

2.2.3 Enterococcus spp. . . . . 6

2.2.4 Campylobacter spp. . . . . 6

2.2.5 Ascaris . . . . 7

2.2.6 Microbial indicators of faecal pollution . . . . 7

2.3 Sanitation systems in Kibera . . . . 7

2.3.1 Open defecation . . . . 8

2.3.2 Pit latrine . . . . 8

2.3.3 Peepoo . . . . 9

2.4 Onsite treatment of human faeces . . . . 9

2.4.1 Primary treatments . . . . 9

2.4.2 Secondary treatments . . . . 10

2.5 Chemical disinfection with ammonia . . . . 11

2.5.1 Urea . . . . 12

2.5.2 Uncharged ammonia speciation . . . . 13

2.5.3 Temperature dependency . . . . 14

2.6 Risk assessment . . . . 14

2.6.1 Sanitation safety planning . . . . 15

2.6.2 Quantitative microbial risk assessment . . . . 17

3 Methodology 21 3.1 Urea treatment . . . . 21

3.1.1 Material . . . . 21

3.1.2 Experimental setup . . . . 21

3.1.3 Sampling and analysis . . . . 22

3.1.4 Statistical analysis . . . . 23

3.2 Risk assessment . . . . 24

3.2.1 Description of sanitation system . . . . 24

3.2.2 Identify exposed groups and exposure routes . . . . 24

3.2.3 Hazard identification . . . . 25

3.2.4 Sampling and analysis . . . . 25

3.2.5 Exposure assessment . . . . 25

3.2.6 Dose-response assessment . . . . 26

3.2.7 Risk characterization . . . . 27

3.2.8 Sensitivity analysis . . . . 28

4 Results 29 4.1 Inactivation rates . . . . 29

4.1.1 Campylobacter spp. . . . . 29

4.1.2 E. coli . . . . 31

4.1.3 Enterococcus spp. . . . . 33

4.1.4 TAN, free ammonia and pH . . . . 35

4.2 Risk assessment . . . . 37

4.2.1 Peepoo sanitation system . . . . 37

4.2.2 Hazard identification . . . . 38

4.2.3 Risk characterisation . . . . 39

4.2.4 Sensitivity analysis @RISK . . . . 44

5 Discussion 46 5.1 Inactivation rates . . . . 46

5.1.1 Campylobacter spp. . . . . 46

5.1.2 E. coli . . . . 47

5.1.3 Enterococcus spp. . . . . 48

5.1.4 TAN and pH . . . . 50

5.1.5 Peepoo toilet as a sanitisation solution . . . . 51

5.2 Risk assessment . . . . 51

5.2.1 Peepoo sanitation system . . . . 51

5.2.2 Exposure assessment . . . . 52

5.2.3 Dose-respons assessment . . . . 53

5.2.4 Risk characterisation . . . . 54

5.3 Sensitivity analysis . . . . 58

5.3.1 E. coli . . . . 58

5.3.2 Ascaris . . . . 58

6 Conclusion 59 6.1 Inactivation through ammonia sanitisation . . . . 59

6.2 Risk assessment . . . . 59

References 60 Appendix 68 A Exposure routes and exposed groups 68 A.1 Schpol A . . . . 68

A.2 School B . . . . 70

A.3 School C . . . . 72

A.4 School D . . . . 74

1 INTRODUCTION

In 2015, 2.3 billion people in the world did not have access to basic sanitation facilities such as improved latrines or toilets, and among them almost 900 million people defecate in the open (9 of 10 live in rural area) (WHO & UNICEF, 2017). In most of Africa, less than 50 % of the population use basic sanitation services. To ensure access to water and sanitation for all is the number 6 of the Sustainable Development Goals (SDG), but progress need to accelerate in order to achieve this goal until 2030. Between 2000 and 2015, in 20 countries, access to the basic sanitation was actually decreasing. With basic sanitation means that there is access to improved facilities that are not shared with other households, but the excreta is still not safely managed. With improved facilites means that it separates human excreta from human contact. Pit latrines without a slab and bucket is an example of unimproved sanitation (Naughton & Mihelcic, 2017).

The WHO & UNICEF (2017) definition of safely managed sanitation service is, beside basic sanitation, that excreta are safely disposed in situ or treated offsite. In many areas this safely management of sanitation is not achieved, due to financial limitations (WHO

& UNICEF, 2017). Because of this, in many urban centres in low- and middle-income countries, large quantities of untreated or partially treated wastewater are discharged into the environment (Evans et al., 2012). When no treatment of the excreta is done, sanita- tion systems can be a source of pollution to surface waters and groundwater. In case of flooding, the pollutants can also spread from the insufficient sanitation system.

Unsafe water, sanitation and hygiene are a major risk to health and can be the cause of diarrhoea which according to WHO (2017) kills 525 000 children under five years old each year, and are the second leading cause of death in the same group. From a systematic review it has been found that most studies determines that sanitation leads to a 30 - 40 % reduction in diarrhoea diseases (Wolf et al., 2014). A safe sanitation system is a system designed to separate human excreta from humans all the way from usage to final disposal or end use (WHO, 2018a) which is desirable not only to prevent infection and diseases but also to ensure privacy and dignity since safe sanitation is recognised as a basic human right. Provide safe sanitation is important in order to decrease the inequality in the world WHO & UNICEF (2017). Not only is it a question about inequality between countries or within, but also between gender. Women with no or inadequate access to sanitation facilities are at high risk when they are forced to use a public toilet or defecate in the open air as they are targets for rape and sexual harassment (Wendland et al., 2017). The lack of privacy and sanitary space is also a problem during their menstrual cycle, which is a rea- son why it is not uncommon that girls quit school when there is no access to school toilet.

Another importance of safe sanitation is that products from a sanitation system could be used either for food production (as a fertilizer) or energy generation. Today considerably large volumes of only partially treated or untreated wastewater are used in agriculture (WHO, 2016). In order to achieve a safe reuse, health issues and aspects of risk reduction is crucial (Stenstr¨om et al., 2011).

The risk of infection when exposed to human excreta is important to consider. Human

faeces may contain a range of pathogenic organisms such as bacteria, parasitic protozoa

and helminths, especially in low-income countries (Stenstr¨om et al., 2011). To reduce hu-

man exposure to pathogens, one of the key goals is safe sanitation (Naughton & Mihelcic, 2017). One way to prevent spreading of diseases is to create a sanitation solution directly after defecation. For this, there are several sanitisation methods available. To achieve safe sanitation for all in a nearby future, the first step is to provide high-quality shared sanitation rather than advanced and expensive technologies (WHO & UNICEF, 2017).

Examples on simple solutions are enclosed collection, single chamber septic tank and storage (meaning latrines not connected to groundwater). Using ammonia as a disinfec- tant have been introduced in order to provide a sanitation solution that could be available to almost all people but that also treats the faeces and creates a product that can be safely used as a fertiliser (Vinner˚as et al., 2009).

Treatment of faeces may prevent transmission of pathogens and other contaminants, but the sanitation system includes several steps such as use, collection, transport (if not in situ), storage or treatment, and disposal or re-use (WHO, 2018a). When designing a san- itation system, exposure routes pathogens to humans have to be taken into account at all steps (Stenstr¨om et al., 2011). In order to do so, there are different tools to predict flows of faeces and to estimate the risk of exposure. Two examples are Sanitation Safety Planning (SSP) and Quantitative microbial risk assessment (QMRA). SSP is a health risk assessment approach to identify health hazards and exposure pathways along the sani- tation chain and to suggest improvements. It does not include any quantification of the risks as QMRA does. QMRA is health risk assessment frameworks that quantifies the risk of infection or illness and are supported by the World Health Organisation (WHO) and increasingly applied in low-income countries (Mills et al., 2018).

1.1 PROBLEM DESCRIPTION

In low income countries, a third of the urban population is estimated to be living in slums (Schouten & Mathenge, 2010). Due to lack of waste och sanitation systems, people liv- ing in slums are at high risk for infection of sanitation related diseases. In Kenya, the prevalence of diarrhoea in slums among children under 5 years old has been estimated up to 32 % which is more than twice as high as in Nairobi and national average (Wambui Kimani-Murage & Ngindu, 2007).

Kibera, an urban slum in Nairobi, is said to be the most densely populated settlement on the continent, with more than 500 000 residents (Stenstr¨om et al., 2011). Most adult residents use simple pit latrines, which are shared by many households, while up to 30 % of the children defecate in the open (Wambui Kimani-Murage & Ngindu, 2007). Parts of Kibera can not be accessed by vehicles, meaning that the emptying of the pits has to be done manually. The workers are extremely exposed and the untreated faeces is emptied into the sewer system, dumped in a stream or disposed elsewhere (Stenstr¨om et al., 2011).

This makes a demand of another solution. Because of this, the Peepoo toilet concept

was introduced (Vinner˚as et al., 2009). The Peepoo toilet is a single-use sanitation solu-

tion made of biodegradable material and uses ammonia as disinfectant. Today, Peepoo

is used in schools in Kibera. Stool samples from children in those schools have shown

a high prevalence of pathogenic organisms, such as roundworms (Ascaris), whip worms

(Trichuris), pathogenic E. coli and Campylobacter spp. ¹ . A high inactivation efficiency is required to assure a safe end-product that could be used as a fertiliser. Furthermore, it is of importance to minimise the direct exposure risks for both workers and users of the Peepoo toilet, and also indirect exposure risks for the environment and the community.

1.2 OBJECTIVE

This study aim to identify and assess the risk of human exposure to pathogens in human excreta when using the Peepoo toilet. This was made by verifying the inactivation ef- ficiency for the ammonia sanitisation that is used in the Peepoo sanitation system, and investigating how the hygienisation could be optimised in order to produce a safe end product. To support a safely managed sanitation service of the Peepoo toilet in Kibera, a risk assessment was performed to point out the present hazards and exposure routes in the Peepoo sanitation system.

Specific research objectives addressed were:

− In what time span are E. coli, Campylobacter ssp., Salmonella ssp. and Enterococ- cus ssp. inactivated in human faeces, when subjected to different temperatures (15, room temperature and 25 ^◦ C) and ammonia concentrations (0%, 1.86% and 3.75%

added urea)?

− How do the inactivation rate of E. coli, Campylobacter ssp., Salmonella ssp. and Enterococcus ssp. in human faeces, correlate to temperature and free ammonia concentration?

− Identify hazards and hazardous events for exposure groups along the Peepoo man- agement chain.

− Estimate the risk of infection by performing a Quantitative Microbial Risk As- sessment (QMRA), presenting the risk for infection of E. coli and Ascaris for the Peepoo management chain.

2 BACKGROUND

This section will provide an overview from the literature and will treat following aspects:

diseases related to human excreta and its way of transmission, sanitation systems in the area of study, basic sanitisation methods, chemical disinfection with ammonia, risk as- sessment and the framework of quantitative microbial risk assessment.

2.1 DISEASE TRANSMISSION FROM HUMAN EXCRETA

Poor sanitation poses a risk of disease transmission where excreta can be a source of enteric pathogens such as bacteria, viruses and parasitic worms. These can spread in the environment through environmental pathways. If the excreted pathogen is ingested by another human it can cause a new infection, via the so called faecal-oral route (Feachem et al., 1983), which is one of several infection routes. The F-diagram, shown in Figure 1, if often used when describing transmission routes of pathogens and barriers that can

1 Nordin, Annika; Researcher at the Department of Energy and Technology; Swedish Argicultural Sci-

ence. Unpublished data of stool samples Kibera, 2016.

prevent transmission of excreta related pathogens. It visualises how pathogens spread via different pathways before reaching a new host, where the importance of different transmission routes varies in different geographic areas (Wanger & Lanoix, 1958). When touching faeces-contaminated areas pathogens are transferred via fingers and infect the host when in contact with the mouth, nose or food, where infectious pathogens can spread through raw or badly cooked food. Contamination from Fields becomes a pathway for transmission via crops, floors or soil and when contaminated water are used for irrigation or faecal sludge is used for fertilizing crops. Fluids like drinking-water and surface water used for bathing composes a risk when contaminated with pathogens, and flies can act as vectors for pathogens and transmit pathogens on their bodies (WHO, 2018a). The F- diagram also describes barriers that can prevent excreta infections (Wanger & Lanoix, 1958) and to achieve safe sanitation the goal is to prevent and stop the exposure pathways through these barriers (WHO, 2018a).

Figure 1. The F-diagram with routes of transmission for pathogens from human excreta and barriers preventing disease transmission (Nordin, 2006)

Treatment is one of the barriers preventing transmission pathways and exposure, which are described in WHO (2018a) Guidelines for sanitation and health. These guidelines ex- plain transmission pathways related to human excreta due to unsafe sanitation, sanitation hazards and hazards events and how they interrelate and eventually expose a new host.

Table 1 describes different hazards, which may occur due to inadequate management in

the sanitation chain, and how these are connected to different hazardous events. The haz-

ardous events are where excreta enters the environment and later expose a new host, and

correspond to the F-diagram (WHO, 2018a).

Table 1. Description of sanitation hazards, inadequate management and hazardous events and their health impact due to unsafe sanitation (WHO, 2018a)

Unsafe toilets or non-existing toilets

Open defecation Insufficient pit toilets

Fields Water Animal contact Vectors

Feet and crops Contact and consumption Further exposure Food, surfaces, fingers Unsafe storage or treatment

Insufficient containment Leakage to groundwater Contact and consumption

Unsafe transportation

Insufficient emptying practices Discharge of untreated excreta

Exposed workers Water

Surfaces and direct exposure Food, water consumption, fingers, feet/skin, surfaces Unsafe offsite treatment

Insufficient treatment

Leakage to groundwater Animal contact

Crops, water consumption Further exposure

Unsafe end use or disposal Untreated faecal sludge

Leakage to groundwater from fertilisation Vectors, animals, water, fields, ground water Crops, water consumption Crops, food, water consumption, fingers, skin, surfaces

2.2 STUDIED ENTERIC PATHOGENS

The environment constitutes the habitat for different microorganisms like bacteria, viruses and parasites. Some of them interact with a host, causing diseases, so called pathogens.

This chapter will cover some enteric pathogens, meaning they are gastrointestinal disease causing microorganisms transmitted via the faecal-oral route (Aw, 2018). Bacteria are single celled organisms found in three different shapes with a length of 0.2 - 2 µm. Some pthogenic bacteria can produce toxin which give symptoms associated with the disease the bacteria causes. Some species are resistant in the environment, and can survive a long time under different conditions (Haas et al., 2014). Some species are capable of growing and multiply outside the host and zoonotic bacteria are able to infect both humans and animals (Ashbolt et al., 2001).

2.2.1 Salmonella spp.

The genus Salmonella spp. consist of two species; Salmonella enterica and Salmonella

bongori. Infections of Salmonella spp., most commonly causing intestinal diseases, is

hard to control due to its high tolerance to environmental stress, resistance, adaptability

and world wide distribution (Chen et al., 2013). Salmonella enterica is associated with

food-borne illness in humans and the infection is caused by ingestion of contaminated

food or water, and the bacteria is excreted in faeces after infection. Salmonella have sev-

eral symptoms, from gastroenteritis to severe illness, where diarrhoea is the most common

disease (Haas et al., 2014). Salmonella (non-typhoidal) is one factor causing diarrhoeal

diseases globally, and can be life threatening for vulnerable groups (WHO, 2018c). Other

strains of Salmonella spp. are Typhi and Paratyphi which cause enteric fever in humans

(Chen et al., 2013) and between 128 000 and 161 000 people die from it every year (WHO,

2018d).

2.2.2 E. coli

Escherichia coli is an adaptive bacteria comprising both non-pathogenic and pathogenic variants (Haas et al., 2014). Even though most strains of E. coli does not cause ill- ness there are pathogenic strains (Harwood et al., 2017), like Enterohaemorrhagic E.

coli (EHEC) producing Shiga toxins. EHEC can lead to life threatning diseases and the serotype O157:H7 is one of the most common serotypes causing outbreakes (Garcia- Aljaro et al., 2017). Other pathogenic strains are Enterotoxigenic E. coli (ETEC) and Enteropathogenic E. coli (EPEC), which are the leading bacterial cause of diarrhoea in developing countries and particularly for children where the illness often is fatal (Har- wood et al., 2017). E. coli is ubiquitous in the normal intestine of humans and animals, but also in other extraintestinal environments like fecal-contaminated areas or polluted waters. It is a common indicator organism for faecal pollution and is often present in a high amount when excreta is present. E. coli spreads through the faecal-oral route and the transmission are through several different routes: ingestion of contaminated food or water, person to person or animal to person contact (Garcia-Aljaro et al., 2017).

2.2.3 Enterococcus spp.

There are over forty species of Enterococcus spp. with various habitats like animals, soil, waters and plants, but it is commonly found in the human and animal intestine. Infection caused by Enterococcus spp. are often due to multi-drug resistance and antibiotic resis- tance of Enterococcus are therefore a leading cause of nosocomial infections. E. faecalis and E. faecium are two of the most common species found in human faeces. Entero- coccus ssp. are widely spread in the environment due to their resistance and survival in different temperatures and pH (Lebreton et al., 2014). Infections caused by Enterococcus ssp. can be difficult to treat (Martin & McFerran, 2017), and since it is an opportunistic pathogen it can become pathogenic when the immune system of a host is impaired (Har- wood et al., 2017). The use of Enterococcus ssp. as an indicator of faecal contamination has been debated since their adaptability to substrates, growth and survival in different environments. There are many non-faecal species of Enterococcus ssp. originating from plants or soil (Lebreton et al., 2014), which also questions its suitability as an indicator of faecal contamination. It is difficult to distinguish the intestinal species from the one natural occurring in the environment. However, when using Enterococcus ssp. as an indi- cator of hand hygiene the origin is believed to be faecal (Boehm & Sassoubre, 2014). It has a higher survival in the environment than viruses and parasites like Ascaris and could therefore function as a model for these organisms (Vinner˚as et al., 2017).

2.2.4 Campylobacter spp.

Campylobacter is one of the worlds most common bacterial cause of diarrhoea and has

serious sequelae. Campylobacteriosis, one of the most common bacterial infection of hu-

mans, was 2010 globally estimated of causing 7.5 million disability-adjusted life years

(DALYs). The many species of Campylobacter spp. has different characteristics, which

makes it hard to understand mechanisms effecting their survival in the environment. Fac-

tors like low temperature and the absence of sunlight are known to be favourably by this

pathogen (Pitkanen & Hanninen, 2017). As a zoonotic bacteria, a variety of animals can

act as reservoirs which makes the source and transmission routes for this bacteria compli-

cated. Infections are mainly food- or waterborne (Pitkanen & Hanninen, 2017).

2.2.5 Ascaris

Ascaris is one of the larger parasitic roundworms and is especially prevalent in humid and warm climates and are common in undeveloped countries where poverty and poor sanitation prevail. It has a high persistence and can survive in the environment for more than six years (Asaolu & Ofoezie, 2018). Ascariasis is an illness caused by Ascaris lum- bricoides and is an infection in the intestine. Both unfertilised an fertilised eggs are found in faeces where the unfertilised eggs are not infective. To get infectious the egg must develop in the environment for 15 - 35 days, and later be ingested by the host. 85 % of the infected people show no symptoms which are common when infected with a small num- ber of worms, while a more heavy infection shows different symptoms depending on the affected body part. Larvae migrating to lungs cause cough, fever and skin rashes and in worse cases asthma, substernal pain, shortness of breath and wheezing. Mild infections in the intestine by adult worms can cause vomiting, nausea, diarrhoea, nutrient loss to name a few, and for a more heavy infection malnutrition, pain, constipation, weight loss and other symptoms are common. The transmission of Ascaris are mainly through the faecal-oral route where the ingestion of the eggs are due to food contaminated by faeces or soiled hands (Asaolu & Ofoezie, 2018). 890 million people are annually infected with Ascaris often due to inadequate sanitation (WHO, 2018a).

2.2.6 Microbial indicators of faecal pollution

Faecal indicator bacteria (FIB) are naturally occurring in the human and animal intestine and are released into the environment through faeces, sludge and other types of faecal waste. Because of their high concentration in faeces, they are used as a indicator of faecal contamination in the environment. FIB are ideally non-pathogenic bacteria and two com- monly used indicators are E. coli and Enterococcus spp.. However there are pathogenic strains of E. coli and some strains of Enterococcus are opportunistic pathogens, such as Enterococcus faecium. Even though these bacteria are used as indicators, they can still survive and grow in the environment, which are to consider when using these indicators in health related investigations (Harwood et al., 2017). Microbial indicators are important tools in health-related investigations and can be used for system assessment, surveillance and monitoring, verification and validation (Farnleitner & Blanch, 2017). For drinking water coliform bacteria and in particular E. coli, is used to detect of faecal contamination.

Still, WHO, EU and US also recommend Enterococcus ssp. as an indicator for recre- ational water quality and for the risk of swimmer illness (Boehm & Sassoubre, 2014).

Faecal indicators are therefore a support for health-risk assessment, and when preforming an QMRA dealing with faecal hazards the use of FIB are essential. In water safety man- agement, using data from epidemiological studies combined with the data from microbial indicators is a fundamental part in microbial diagnostics (Farnleitner & Blanch, 2017).

Both E. coli and Enterococcus ssp. has recently been used as indicators for hand hygiene, where Enterococcus ssp. is preferable due to its correlation to hygiene indicators and its traceability to specific activities (Boehm & Sassoubre, 2014).

2.3 SANITATION SYSTEMS IN KIBERA

Safe sanitation is of great importance for human health, well being and to prevent infec-

tious organisms from spreading in the environment. Sanitation is defined as the provision

and access to services and facilities for safe disposal of human excreta. The system man-

agement should be safe throughout the whole chain, from the toilet to the storage includ- ing step like transport, treatment and end use (WHO, 2018a). In 2015 39 % of the world population did not use a safe sanitation solution, and today 2.3 billion people still don’t have sanitation facilities i.e toilets or improved latrines (WHO., 2018b). In Kibera, the sanitation system are mostly on site. According to Worrell et al. (2016), the most common method of excreta disposal is shared sanitation facilities (81.7 %). Among those, more than half of the households shared their facility with 10 other households or more. 1.3 % of the households had access to improved sanitation facilities and 13.9 % to unimproved sanitation facilities. 3.1 % of household practised open defecation, meaning having no ac- cess to any sanitation facilities. It was also reported that more household members were using open defecation during night time (Worrell et al., 2016). In some of the schools also the Peepoo sanitation system are used. This section will further describe these ways of excreta disposal and their related ways of exposures.

2.3.1 Open defecation

Open defecation is according to (Stenstr¨om et al., 2011) not a part of the sanitary system but is a common way of disposing excreta in developing countries. In places like urban slums where adequate toilet solutions are missing flying toilets are common, meaning excreta are put in a bag or similar and then thrown away. Included in open defecation is also cat latrine, which are when excreta are disposed in a shallow hole and then buried close to the surface, and open latrines, which are when excreta not are covered. In densely populated areas, such as Kibera, the likelihood of exposure to fecal material is high and the children are the must vulnerable exposure group since they have a higher frequency of contact with contaminated soils than adults. In these areas, storm water and surface water have an impact on the exposure to pathogens. Other transmission routes that are considered for open defecation, except through direct contact, are contamination of crops, soil and ground water. The considered risks groups for open defecation are the user and the community and for them, the level of risks of exposure to pathogens with fecal origin is high. The exposure frequency is also considered high for both the users and for the community who lives nearby the contaminated area or pass by (Stenstr¨om et al., 2011).

2.3.2 Pit latrine

A pit latrine is a shaft dug into the earth, which can be supported with reinforcing material to not collapse (Stenstr¨om et al., 2011). It is not connected to any sewer system and needs to be emptied. This is mostly done manually in Kibera, posing risk for the worker of exposure to excreta with high concentrations of pathogens. The frequency of the emptying depends on the design of the pit. A pit latrine risks polluting the ground water if the ground water level is high, and also serves as a source of surface water contamination during floods. Flies breeding in the pits is also considered as a route of transmission. The reduction of pathogens in pit latrines are higher than in bucket latrines (which is further described in the section below), and it could be enhanced by addition of lime or ash.

For the user, the frequency of exposure is low since the only exposure way for the user

is flies. Also for the community, the frequency of exposure is said to be low since it is

only potential affected by groundwater or surface water. The total level of risk however

is considered medium for both user and community due to the additive effect. To reduce

the risk of exposure through flies, a Ventilated Improved Pit latrine (VIP) can be used, in

which a ventilation pipe is installed into the pit. The ventilation pipe is used to control flies and exhaust the odour from the pit (Stenstr¨om et al., 2011). However, the health risks from flies are not completely removed by ventilation.

2.3.3 Peepoo

Stenstr¨om et al. (2011) describes the Peepoo solution as a slightly better variant of open defecation. The Peepoo is a bag made of biodegradable material that is opened and put into a small potty, called kiti (Vinner˚as et al., 2009). It could also be held by hand or put into a small bucket. The user defecate into the bag and seals it with a simple knot.

Inside the bag there is a small package of 6 grams urea (CO(NH ₂ ) ₂ ). When urea comes in contact with the faeces, bacterial enzymes degrades it and it works as a sanitizing agent (Vinner˚as et al., 2009). This is described further in Section 2.5.

2.4 ONSITE TREATMENT OF HUMAN FAECES

In order to reduce pathogens in human excreta, different sanitisation methods are used.

Regardless method, it is not possible to achieve a complete die-off, but the aim is to reach a low level of pathogenic microorganism so that the end product don’t cause infection. The reduction of pathogens by treatments often follows a log linear relationship, why time to reduce a tenfold of the concentration (the D-value) is often used in order to measure the efficiency of the technology (Vinner˚as et al., 2017). Guidelines from WHO (2006a) for verification of faecal treatments are shown in Table 2.

Table 2. Guideline values for verification of faecal treatment both as allowed concentra- tion in the end-product and as efficiency of treatment (the required log ₁₀ reduction).

Guideline Helminth eggs E. coli Pathogens

WHO (2006a) < 1/g total solids < 1000 g /solids 6 log ₁₀ red. ^a

a Based on excreta storage without fresh additions. For fresh faeces, the guideline is 8 - 9 log 10 reduction (2 log 10 higher than for wastewater). For source-separated urine it is 3 - 5 log 10 units (WHO, 2006a).

2.4.1 Primary treatments

A primary treatment is a treatment that can be achieved directly into the toilet. This section will briefly present some treatment technologies that can be used in low-income countries. Chemical disinfection with ammonia is not included since it is described in detail in Section 2.5.

Prolonged storage

If no treatment is applied to the faeces, it is recommend to be stored for more than a year (at a temperature between 20 and 35 ^◦ C) in order to achieve more or less complete inacti- vation of Ascaris eggs. The faeces should be stored in a closed environment and no fresh excreta should be added during the period of storage (WHO, 2006a).

Chemical disinfection: Alkaline treatment

If adding lime or ash after each defecation, an increased pH is obtained which has a sani-

tising effect. However, alkaline treatment may not achieve a complete elimination but a

substantial reduction (WHO, 2006a). If a pH above 9 is reached, the time of treatment should be at least 6 months. It has been observed that a pH above 12 is required to inac- tivate Ascaris eggs within 3 months (Eriksen et al., 1996). To achieve such a high pH, a large amount of ash may have to be added (Boost & Poon, 1998).

Drying faeces

When using desiccating toilets, also known as dehydrating or drying toilets, the urine is preferably separated from the faeces to achieve a moisture content below 30 - 40 % (Nordin, 2010). It is made in a basic construction with a chamber below the toilet (WHO, 2006a). Drying the faeces is desirable for inactivate microorganisms, and rising tempera- ture also speed up the process. Ascaris eggs has been shown to tolerate moisture content as low as 5 % (Feachem et al., 1983). The treatment works best in arid environments since dry air can be circulated into the toilet chamber to remove evaporated moisture (Nordin, 2010). Adding soil, lime or ash can also lower the moisture content and increase pH which enhance the pathogen die-off (WHO, 2006a).

Compost

Composting toilets have the same construction as the desiccating toilets though it can be designed with or without urine diversion (WHO, 2006a). It relies on aerobic digestion of organic matter and the sanitisation effect is achieved by high temperatures. If the compost is thermophilic, the reduction of pathogens may be fast and the minimum requirement of treatment is then 1 week. However, this temperatures are normally not reached when composting is used as a primary treatment. The temperature range in the compost are normally mesophilic or ambient which leads to a slower inactivation. Fresh faeces are also normally too wet and exhibits a too low carbon nitrogen ratio, why dry, carbon-rich bulking material has to be added. To preserve the aerobic conditions, the moisture of the content should not exceed 60 % and in order to ensure that all content reach high temper- atures, a thorough mixing of the material during the high active phase of composting is needed (2 - 3 times per week for two weeks) (Niwagaba, 2009). Since the process may be difficult to run, compost is normally used as a off-site secondary treatment (WHO, 2006a).

2.4.2 Secondary treatments

Faecal material collected from latrines and pits may contain high concentrations of pathogens.

Therefor, a secondary treatment may be needed to reach below guideline levels of pathogen content (WHO, 2006a). This section present some alternatives of low-cost secondary treatments that is conducted after collection of the faecal material. As concluded above, thermophilic composting is an option for secondary treatment and will not be described further.

Anaerobic digestion

Anaerobic digestion is a biological process without access to oxygen that can convert organic substrates to biogas. To be beneficial for pathogen inactivation, a thermophilic temperature (> 45 ^◦ C) and prolonged hydraulic retention time is needed (Nordin, 2010).

Most often, small-scale digestion are performed at ambient or mesophilic temperatures

resulting in limited pathogen reduction for which the process demands monitoring and

external energy. To achieve hygienic quality requirements, a post-treatment is needed (WHO, 2006a).

Incineration

Combustion of faeces is a compact and rapid process that destroys pathogens. The only end-product is ashes which can be re-used as a cover material in primary treatments.

However incineration decreases 70 - 90 % of the organic matter, total nitrogen and plant- available phosphorous. To prevent smell and smoke development, the faecal material also has to be dried to about 10 % moisture content prior to incineration wherefore it is of limited interest as a low-tech sanitation method (Niwagaba, 2009).

2.5 CHEMICAL DISINFECTION WITH AMMONIA

To inactive pathogens in human excreta one treatment option is chemical disinfection with ammonia. The technique is simple and can be used at a low cost (Fidjeland et al., 2015).

It has been showed to be more efficient for attaining hygienically safe faecal matter than other sanitisation techniques such as composting and storage, and the added ammonia do not only sanitize the treated material, but also gives it an increased fertilizer value (Vin- ner˚as, 2007). Also, mesophilically digested sewage sludge treated with ammonia have showed lower greenhouse gas emissions (N ₂ O and CH ₄ ), than if stored untreated. This probably is due to inhibition of the microbes that produces these gases (Will´en et al., 2016) .

In this study total ammonia nitrogen (TAN) refers to NH-N, which includes uncharged ammonia (NH ₃ ) and ammonium ion (NH ⁺ ₄ ). Ammonia is a natural product of urea hydrol- ysis (further described in Section 2.5.1) and of ammonification (the microbial degradation of nitrogen-containing compounds such as proteins). Ammonia is one of the products of anaerobic digestion (a commonly used process for degrading biological waste) of organic compounds rich in nitrogen (Bujoczek, 2001). Uncharged ammonia is long known to be toxic to most cells, and its impact on cell structures is well reported though the mechanics behind it are not yet fully understood (Schneider et al., 1996; Diez-Gonzalez et al., 2000).

However, it is known that free ammonia diffuses across cellular membranes and destroys membrane potential and denatures proteins. The diffusion is believed possible due to the small size of ammonia and its high solubility. When the free ammonia has passed through the membrane it causes a rapid alkalinization of the cytoplasm, and to maintain inter- nal pH, the cell takes up protons from the outside and then looses potassium ions (K ⁺ ), which is an important nutrient and the lack of it causes death of the bacterial cell (Bu- joczek, 2001). However, the hypotheses about the biocidal effect of NH ₃ does not apply for viruses, since viruses do not have a cell metabolism of their own. For viruses, NH 3

could damage components such as protein, envelope or nucleic acid. How persistent the viruses are against ammonia depends on their genome style (Decrey et al., 2016). It could also be that not only ammonia has an inactivating impact in higher pH. For instance, at pH 9.5 other inactivating effect on virus has been observed in buffer solutions by Decrey et al.

(2015), which is suggested to be due to other bases in the background solution, such as

hydroxide, (bi-)carbonate or phosphate. For inactivation of oocyst, however, it has been

showed that pH rather has an indirect effect on oocyst wall permeability and that it is the

chemical activity of free ammonia that is the factor of oocyst inactivation (Jenkins et al.,

1998), and it when it comes to inactivation of Ascaris eggs high pH alone did not cause inactivation, but plays an indirect role of converting the ammonia ion to its uncharged form (Pecson & Nelson, 2005). Other study where virus was inactivated could not isolate the effect of pH and ammonia concentration, only state that the inactivation of viruses at pH 9 and above is both correlated to ammonia concentration as well as to pH (Magri et al., 2015).

It has been shown that the free ammonia, rather than TAN, is causing an inactivation of parasitic protozoa (Jenkins et al., 1998), viruses (Emmoth et al., 2011; Burge et al., 1983), bacteria (Himathongkham et al., 2000; Ottoson et al., 2008) and helminth eggs (Pecson & Nelson, 2005; McKinley et al., 2012), where Ascaris eggs, is said to be the most persistent pathogens present in faeces regarding ammonia inactivation (Fidjeland et al., 2015). Using ammonia as an inactivation method, species has been tested in in waste water sludge (Ghigliettia et al., 1997; Pecson et al., 2007), fecal sludge (Magri et al., 2015), manure (Park & Diez-Gonzalez, 2003; Ottoson et al., 2008), faeces (Magri et al., 2013), hatchery waste (Emmoth et al., 2011), syntetic urine (Oishia et al., 2017) and human urine (Vinner˚as et al., 2008). The technology of ammonia sanitization is simple.

It only needs a source of ammonia (normally aqueous ammonia or urea) and an airtight storage to prevent the uncharged ammonia to be lost to the atmosphere. The ammonia is not consumed during the process, which retains the fertilizer value (Nordin et al., 2009).

Since ammonia exists naturally in urine, the concentration of intrinsic ammonia may be enough in fecal sludge from toilets with low water use, but to ensure inactivation more ammonia in form of urea can be added (Fidjeland et al., 2013). However, to lower the costs as little added urea as possible is desirable. Increased temperature could be another option to make the treatment more efficient (Fidjeland et al., 2015). Since a higher pH shifts the ammonia equilibrium so that the proportion of uncharged ammonia increases (see Section 2.5.2), an alkaline agent such as lime or ash could be used to improve the treatment. In order to lower the costs the storage time should be as short as possible, since the storage facility is a major cost for ammonia sanitization. Therefor, it is important to be able to predict the inactivation time of different pathogens. In conclusion, the main variables that affect the treatment efficiency are temperature, ammonia concentration, and pH (Pecson et al., 2007). Increasing these parameters all leads to a shorter treatment time, as will be described further in following sections (2.5.1 - 2.5.3).

2.5.1 Urea

Even though aqueous ammonia solution have a sanitising effect, it is not normally used as a disinfectant in animal manure because of its high volatility and toxicity, which could lead to a harm for the farmers and the environment, and also decrease its value as a fertil- izer (Park & Diez-Gonzalez, 2003). Adding urea instead of aqueous ammonia is safer for the worker (Nordin et al., 2018) and therefor preferable (Ottoson et al., 2008).

Around 80 % of the nitrogen in the urine is in form of urea (CO(NH 2 ) 2 ), and the rest of the nitrogen is found in the form of ammonium/ammonia and creatinine (J¨onsson et al., 2000). The hydrolysis of urea is catalyzed by the Ni(II)-containing enzyme urease, which makes the reaction 10 ¹⁵ times faster than when it is uncatalyzed (Mazzei et al., 2017).

Urease is found in plants, algae, fungi, and several microorganisms (Maroney & Ciurli,

2014). The presence of urease makes urea to decompose when in contact with biological matter such as faeces, see Equation 1. The decomposition of urea leads to an alkaline pH, which normally stabilises at pH 9 (Hellstr¨om et al., 1999).

CO(N H ₂ ) ₂ + 3H ₂ O −−−→ 2N H ^urease ₄ ⁺ + HCO ⁻ ₃ + OH ⁻ (1) Adding urea does not only increase the ammonia concentration but also carbonates which also has been suggested to have sanitising effect when in the ionic form, CO ²⁻ ₃ (Park &

Diez-Gonzalez, 2003). It has been argued that it is carbonate and not ammonia that is the dominant inactivating factor in urea treatment, since adding NH 4 Cl and increasing pH did not give a sanitising affect within 24 hours for organisms such as Escherichia coli O157:H7 , Escherichia coli K-12, Salmonella Typhimurium, along others, in a study made by Diez-Gonzalez et al. (2000), as the adding of Na 2 CO 3 or urine and urease did. But even when ammonia has been confirmed as the dominant inactivating factor, it is important to consider the change in the speciation of carbonate species since it gives an additive effect (Decrey et al., 2015). For lower temperatures, it has been showed that urea has a higher impact on the inactivation rate of Salmonella spp. than ammonia solution, even though the uncharged ammonia concentration was lower in the urea treatment. This indicates that the carbonate are of importance for the inactivation in these cases (Fidjeland, 2015).

When investigating the inactivation of Enterococcus faecalis, adenovirus, reovirus and bacteriophages, Fidjeland (2015) couldn’t separate the effect from high pH or concentra- tion of carbonates or ammonia, since they were all correlated. Carbonates however did not have an inactivating effect on Ascaris eggs. It has been suggested to use carbonate and uncharged ammonia in combination for manure treatment since they are thought to have different killing mechanisms (Park & Diez-Gonzalez, 2003). As for ammonia/ammonium equilibrium, the carbonate equilibrium is pH and temperature-dependent (Nordin, 2010).

2.5.2 Uncharged ammonia speciation

Depending on ammonia concentration and pH, ammonia can function as a fertilizer or as a toxic agent to the microbial community. As seen in Equation 4, the fraction of uncharged toxic ammonia is increasing with increasing pH. This is also understood when observing Equation 2. How NH 3 is distributed between gas and aquatic form depends on temperature and the ratio can be calculated by Henry’s law constant H (Nordin, 2010).

Due to its polarity and ability to form hydrogen bonds, ammonia gas is highly soluble in water. As seen in Equation 2, ammonia acts as a weak base when in solution since it produces hydroxide ions by de-protonation of water. The distribution between uncharged ammonia and ammonium ions can be quantified by the dissociation constant, K a . The pK a between 0 - 50 ^◦ C can be calculated according Equation 3, where T is the temperature in degrees Kelvin. The fraction of NH ₃ in a aqueous solution is calculated according to Equation 4, which shows that the fraction of dissolved ammonia is depending on pH and temperature, increasing with either of them.

N H 3 (gas) ← → N H ^H 3 (aq) + H 2 O(l) ←→ N H ^K

₄ ⁺ (aq) + OH ⁻ (aq) (2)

pK _a = 2729.92/T + 0.090181 (3)

f _{N H3} = 1/(10 ^pK

^−pH + 1) (4)

Uncharged ammonia is starting to prevail at a pH higher than 8 (Bujoczek, 2001) and at pH 11 more than 90 % of the ammonia is present as NH 3 (Nordin, 2010). However, only a slightly alkaline pH is needed to cause microbial inactivation if the amounts of total ammonia are high (Nordin et al., 2009). The effect of ammonia treatments are highly dependent on alkaline pH (Park & Diez-Gonzalez, 2003). A record of pH variations could be used to detect the start and the rate of the decomposition of urea (Hellstr¨om et al., 1999).

2.5.3 Temperature dependency

As was demonstrated in Equation 4 above, not only pH but also temperature has an influ- ence on the NH 3 speciation and thus the inactivation, having a larger impact at moderately alkaline pH (8 - 10) (Nordin, 2010). It has been observed by Decrey et al. (2015) that tem- perature also affects the second order rate constant k _{N H}

of viruses, which was quantified with the Arrhenius relationship, showing linear dependence of ln NH 3 on 1/T. Arrhe- nius equation in its integrated form express the relationship between the rate of reaction constant and temperate as follow:

logk = −E/RT + C (5)

where T is the temperature in degrees Kelvin, E is the energy of activation, R is the molar gas constant and C is the intercept on the log k axis (Burge et al., 1983). For temperatures above 40 ^◦ C a biphasic Arrhenius plot for f2 phages and poliovirus was observed by Burge et al. (1983), which indicates that there are two different inactivation processes: one for low and one for high temperature. Magri et al. (2015) did not observe any clear correlation between temperature and influence of ammonia and pH on viruses, when investigating temperatures between 10 and 28 ^◦ C. Magri et al. (2015) means that there is no increase in the membrane passage or attachment to surface proteins by ammonia with temperature, but that there is an indirect impact on inactivation since higher temperatures leads to a higher fraction of NH 3 . When it comes to helminth eggs, Pecson et al. (2007) showed that the inactivation rate are strongly dependent on temperature. In all cases of that study, each 10 ^◦ C increase in temperate caused a statistically significant reduction of the time for 99 % inactivation (t 99 ), regardless if more free ammonia was added or not.

2.6 RISK ASSESSMENT

To assess the risk of human exposure to different hazards a risk analysis, including risk as- sessment, risk management and risk communication, should be performed (CAC, 1999).

A risk assessment is the characterisation and estimation of health effects for people ex-

posed to hazards. The risk assessment are either quantitative or qualitative, where a quan-

titative assessment numerically estimates probabilities of e.g. infection or illness (CAC,

1999; Haas et al., 2014). The different hazards can be physical, chemical, microbial

agents, material, situations or a combination of the different parts (Haas et al., 2014). In

this study the risk assessment was based on the methodology of WHO (2016) manual San-

itation safety planning (SSP) and the framework of quantitative microbial risk assessment

(QMRA), both further described in Chapter 2.6.1 and 2.6.2.

2.6.1 Sanitation safety planning

WHO’s manual Sanitation safety planning (SSP) focus on safe use of human excreta, and is a risk based management tool for sanitation systems (WHO, 2016). It is targeted to assist the implementation of WHO guidelines for Safe Use of Wastewater, Excreta and Greywater but according to (WHO, 2016) the methods of SSP can be applied to all san- itation systems. The approach of SSP is to systematically identify and manage health risk along the sanitation chain (Mills et al., 2018). To identify health hazards and ex- posure pathways, it uses local knowledge and the likelihood and severity of risks might be subjective. However, it promotes a multi-barrier approach with a focus on achieving pathogen log reduction and there is a step-by-step guidance for performing a SSP. The structure from WHO (2016) is based on six different modules, see below. Each module is divided into different sections with additional guidance notes, tools and examples.

1. Prepare for SSP

2. Describe the sanitation system

3. Identify hazardous events, assessing existing control measures and exposure risks 4. Develop and implement an incremental improvement plan

5. Monitor control measures and verify performance 6. Develop supporting programmes and review plans

The main objective in Module 2 is to generate a description of the sanitation system

with relevant boundaries. To support the following risk assessment an understanding of

the different parts in the sanitation system is necessary, which it the purpose of Module

2. Investment in system monitoring and other improvements has to respond to the risks

related to the highest health impact, which Module 3 ensures. This module also prioritise

the exposure risk by identify hazards and hazardous events and access control measures

(WHO, 2016). In Table 3 a more specific description of Module 2 and 3 are shown.

Table 3. Summary of Module 2 and Module 3 used in WHO’s manual Sanitation safety planning (SSP) (WHO, 2016)

Module Description

2.1 Map the system Create an understanding of path and source of waste through the system

2.2 Characterise the waste fractions Aims to describe constituents from waste sources

2.3 Identify potential exposure Initial classification of exposed groups related to exposure in the sanitation system

2.4 Gather compliance and contex- tual information

Gather information about the context in which the system exist

2.5 Validate the system description Assurance that the system description is complete 3.1 Identify hazards and hazardous

events

A detailed identification of how the risk occurs when the sanitation system is operating

3.2 Refine exposure groups and ex- posure routes

A detailed identification of who are exposed when the sani- tation system is operating

3.3 Identify and assess existing con- trol measures

Describes how the system protects groups of risk 3.4 Assess and prioritize the expo-

sure risk

Show a structure which helps identify and prioritize the highest risk

The method in Module 2 suggest the use of a flow diagram to illustrate the system, de- pending on the complexity of the system. This will help to track the path of the waste from the point of generation all the way to disposal, and SSP provides a guidance note (2.1) to be used as a checklist. A site visit is important to understand the whole sanita- tion chain and the variability of the system (WHO, 2016). Module 2 also give a tool to categorise people exposed to a hazard, see Table 4.

Table 4. Classification of exposure groups (Stenstr¨om et al., 2011).

Group Symbol Short description

Workers (W) Person maintaining, cleaning, operating or emptying the sanita- tion technology.

Farmers (F) Person using the final product from the sanitation solution.

Local community (L) People living near to or downstream from the sanitation technol- ogy or farm on which the sanitized material are used in, and are passively affected.

Consumers (C) People consuming or using products that are produced using the sanitized products

Module 3 explains the difference between hazards and hazardous events and presents

tools to identify hazards and hazardous events for the relevant sanitation system. This

is done alongside the sanitation chain which are described with Module 2. The method

also makes sure different circumstances are taken into account when the identification is

completed. Tools are provided to describes each particular exposure groups from Module

2 in more detail, see Table 5. The purpose of this tool is to evaluate key questions for the

exposed groups to assess the risk. The chapter also provides guidelines how to assess and

prioritize the exposure risk 2.6.1.

Table 5. Key questions to assist identifying and refining exposure groups and exposure routes (Stenstr¨om et al., 2011).

Question Description of question

Who are they? Description of people and what they do in relation to exposure.

Consideration should be given to vulnerable sub-groups consid- ering age, gender and factors of social exclusion.

How many are there? Give actual numbers, if known, otherwise estimate and give basis of estimate. Number of people (individuals) likely to be exposed directly or indirectly

Where are they? Explain where the exposure occurs within the sanitation system to explain how they might be exposed to hazards.

What they are exposed to?

What contaminant and in what circumstances (e.g. chemical, mi- crobial due to barrier failure, extreme weather etc.).

What is the route of contamination?

Infection route to be considered (e.g. through skin, ingestion of crops, soil or water, intermediate vector).

How often are they ex- posed to this?

Exposure frequency. Is it every time, daily, weekly or perhaps just once a year? If do not know, have a “guesstimate”.

What dose? Defines the likely dose of exposure. This depends on the local situation and is sometimes difficult to estimate. The dose will also differ between groups of individuals but an “estimate” is still of value.

2.6.2 Quantitative microbial risk assessment

A tool to predict the risk from a actual or expected exposure to infectious pathogens is Quantitative Microbial Risk Assessment (QMRA). The concept is based on a methodol- ogy from chemical risks developed by National Research Council (1983) and was first establish for drinking water. The method has been used for other purposes and since microorganism differs from chemicals the concept has developed to better describe the assessment for health issues related to microorganisms. The QMRA has four subse- quent steps; hazard identification, exposure assessment, dose-response assessment and risk characterization (WHO, 2006b; Haas et al., 2014; National Research Council, 1983).

The hazard identification determines which pathogens to include, and thereafter an assess- ment are conducted, where the dose of relevant pathogens to which humans are exposed and infected are determined. The dose-response model are used to give a probability of infection, so that finally a risk can be characterized with consideration of the exposure assessment. Before implementing these step the purpose of the risk assessment should be clearly stated. According to CAC (1999) the output form and alternatives should be specified.

Hazard identification

The purpose of hazard identification is to describe the effect a hazard can have to human

health and to identify relevant and representative risks. The hazard identification identifies

human illness and diseases associated with each specific microbial agent for the systems

studied for each specific case (Haas et al., 2014; WHO, 2006b). The infection can be any-

thing from asymptomatic to death (Haas et al., 2014) and the hazard identification also

aims to determine if exposure to a specific microbial agent can increase the number of