Note: This report was originally published in 2015. As of 2019, maternal and neonatal tetanus has been eliminated in 47 out of 59 high-risk countries.1 UNICEF and Kiwanis International, which together have raised more than $100 million aimed at maternal and neonatal tetanus elimination,2 continue to work in the 12 remaining high-risk countries.3 Given the scale of the UNICEF/Kiwanis campaign, at this time we do not expect that this intervention has substantial room for additional funding in the locations at highest risk. Furthermore, the 12 remaining countries may present implementation difficulties due to conflict, remoteness, or a lack of access to healthcare,4 making the intervention more costly per life saved than is represented in our original cost-effectiveness analysis.

Published: April 2015

What is the disease targeted by the program?

Tetanus is a disease characterized by painful muscle contractions.5 An infection results when bacterial spores in the soil and in animal feces contaminate damaged tissue.6 Unhygienic birth and postnatal practices are associated with higher rates of tetanus.7 Without medical care, the risk of death is very high, especially among newborns.8 Over the past two decades, the global number of deaths due to tetanus has substantially declined.9 Cases now cluster in poor and remote areas with less access to health services.10

What is the program?

Immunization with a tetanus toxoid (TT)-containing vaccine can prevent tetanus.11 Babies born to mothers with immunity also can become protected.12

In 1989, the World Health Assembly (WHA) adopted a resolution to "eliminate" neonatal tetanus by 1995.13 The resolution defined "elimination" as the reduction of neonatal tetanus to a district rate of less than 1 case per 1,000 live births worldwide (true elimination is not feasible since tetanus cannot be removed from the environment).14 This initial target date was missed.15 In 1999, WHO, UNICEF and the UN Population Fund partnered to re-launch a global, coordinated effort to eliminate maternal and neonatal tetanus called the Maternal and Neonatal Tetanus Elimination (MNTE) Initiative (http://www.who.int/immunization/diseases/MNTE_initiative/en/index.html).16 Between 2000 and 2013, 34 countries eliminated MNT, as assessed by the MNTE Initiative. In general, the MNTE Initiative validates elimination in a country as follows:

1. A country requests a formal validation of elimination.17
2. The MNTE Initiative reviews available data to identify the highest risk district.18
3. The MNTE Initiative conducts a survey in the highest risk district with a sampling method designed to assess whether that district has a neonatal tetanus mortality rate of less than 1 death per 1,000 live births.19
4. If the survey shows that the district has a neonatal mortality rate of less than 1 death per 1,000 live births, then the MNTE Initiative determines that the country has eliminated maternal and neonatal tetanus.20 We are not aware of any follow-up system for determining if the death rate in a district or country that has "eliminated" MNT later climbs back above one per 1,000 live births.

As of 2013, 25 countries still had not eliminated MNT.21

In this report, we focus on a particular, population-level approach to immunization implemented by the MNTE Initiative. This "high-risk area" approach consists of immunization campaigns with multiple rounds targeting all women of reproductive age in areas at high risk of maternal and neonatal tetanus within a country.22 The campaigns are often called supplemental immunization activities (SIAs), because they supplement any programs that routinely provide immunizations through the health system.23 They are used especially in areas with high barriers to access to health services.24

Though the MNTE Initiative also implements other interventions, such as training programs for skilled birth attendants, the promotion of clean birth practices, and immunization of pregnant women at antenatal clinics; we do not address those here.25

Does the program work?

Strong evidence indicates that babies born to vaccinated women are immune to tetanus for some time.

We have found limited evidence demonstrating that SIAs, specifically, reduce MNT and limited information demonstrating that SIAs target populations with high neonatal tetanus mortality rates, cover their target populations, and successfully immunize the population reached.

Are vaccinated individuals protected?

There appears to be no debate that the TT vaccine is effective, safe, and stable.26 It has been widely used in the developed world for decades.27 The initial studies underlying its wide use demonstrated the presence of antibodies in immunized human subjects at levels required for immunity, as inferred from animal models.28 Because its efficacy appears to be widely accepted, we have not vetted this evidence carefully.

Are the babies born to vaccinated women protected?

Strong evidence indicates that babies born to vaccinated women are immune to tetanus for some time. Demicheli, Barale and Rivetti 2013, a Cochrane review, includes two studies reporting the effect of TT immunization of women on neonatal mortality and concludes that maternal immunization protects newborns:29

Study	Study type	Sample size (infants)	Location	Effect	Interval between immunization and birth	Outcome
Newell et al. 1966	Randomized controlled trial (RCT)30	1,18231	Colombia32	1 dose: 43% (95% CI: -24% to 74%)33 2-3 doses (mostly 3): 98% (95% CI: 70%-100%)34	0-54 months35	Reduction in neonatal tetanus mortality36
Black, Huber and Curlin 1980	Quasi-RCT37	8,64138	Bangladesh39	1 dose: 46% (95% CI: 11% to 68%)40 2 doses: 68% (95% CI: 50% to 79%)41	9-32 months42	Reduction in 4-14 days neonatal all-cause mortality43

What immunization programs did the trials examine?

Newell et al. 1966 randomized 2,776 women between the ages of 13 and 45 to a treatment group or a control group.44 Of these women, Newell et al. 1966 excluded 1,158 from analysis, because they declined any immunization.45 The study offered the remaining 772 women in the treatment group 3 doses of a tetanus toxoid-containing vaccine, each spaced at least 6 weeks apart.46 All these women received at least 1 dose, 67% received at least 2 doses, and 48% received 3 doses.47 The study does not describe why it did not achieve higher coverage rates for the second and third doses.

Black, Huber and Curlin 1980 immunized 46,443 children between the ages of 1 and 14 years and non-pregnant women at least 15 years old and then attempted to immunize the same population 6 weeks later in a second round in which 74% received a second dose.48

Do SIAs reduce MNT?

Even with an efficacious vaccine, immunization campaigns may face a number of challenges in delivering vaccines to the appropriate populations.

We have seen limited evidence examining the impact of SIAs on tetanus mortality. Of the 2 randomized or quasi-randomized studies and the 7 non-randomized studies that Demicheli, Barale and Rivetti 2013 identifies, 1 quasi-RCT and 2 non-randomized studies report on the effect of a mass immunization campaign on tetanus mortality:49

• We discuss Black, Huber and Curlin 1980 above.
• Yusuf et al. 1991 examined surveys before and after an immunization campaign in the Pidie district in Indonesia and found a large reduction in neonatal tetanus mortality.50 It also found that the campaign achieved a high coverage rate, but we are unsure if the coverage rate was measured in the surveys or came from administrative reports, which may overestimate coverage. We have not reviewed this study carefully because we could not find it online, and chose not to try to obtain it.51
• Mass immunization campaigns took place in high-risk areas in China in 1995-1996.52 Chai et al. 2004 found that neonatal tetanus cases per 1,000 live births reported by the national surveillance system increased from 0.18 in 1996 to 0.21 in 1997 and then declined from 0.20 in 1998 to 0.16 in 2001.53 However, we are not confident in the accuracy of the surveillance system.54 The study also reported that SIAs achieved high coverage rates, but we are unsure of the source of this data.55

We have found only limited information answering the questions below:

• Do SIAs target populations with high neonatal tetanus mortality rates?
• Do SIAs cover their target population?
• Do SIAs successfully immunize the population reached?

What data do we rely on?

The MNTE Initiative has published reports on progress towards and validation of elimination in various countries.56 These reports discuss two types of surveys: (1) "assessment surveys" conducted to assess neonatal tetanus mortality in an area and (2) "validation surveys" conducted to validate elimination of tetanus in an area. Countries request a validation survey when they believe they have eliminated tetanus.57 It is unclear how the MNTE Initiative chooses the countries in which to conduct assessment surveys. The assessment surveys provide a precise estimate of mortality, while the validation surveys only determine whether or not the mortality rate exceeds the threshold for elimination.58

We discuss these reports in more detail below though the areas surveyed may differ from areas typically targeted for SIA in important ways. We have found limited other information that directly addresses the questions above.

Do SIAs target populations with high neonatal tetanus mortality rates?

We have limited information about whether the MNTE Initiative successfully targets areas with high rates of neonatal tetanus mortality.

The MNTE Initiative aims to target areas with more than 1 neonatal tetanus case per 1,000 live births, or about 1 neonatal tetanus death per 1,000 live births (without hospital care, the case fatality rate is very high).59 We have not seen baseline surveys that could confirm that the MNTE Initiative effectively targets areas with high neonatal mortality rates in practice.

Five of the MNTE Initiative reports describe surveys that measure neonatal tetanus mortality rates ("assessment surveys").60 These surveys reported neonatal tetanus mortality rates ranging from 2 deaths per 1,000 live births to 8.6 per 1,000.

Table: Surveys assessing neonatal tetanus mortality rates

Country	Location	Year	Neonatal tetanus death rate per 1000 live births (95% confidence interval)
Nigeria	Kano61	200662	5.9 (2.8-8.9)63
Afghanistan	Badakshan, Kandahar, Kunduz64	200465	Badakshan: 5.0 (1.9-8.1), Kandahar: 2.5 (0.2-4.7), Kunduz: 4.1 (1.1-7.0)66
Lao People's Democratic Republic	9 high-risk districts in 3 provinces67	200168	8.6 (4.2-13.1)69
Niger	Birni N'Konni, Keita, Mirriah70	200171	2 (0-5.2)72
China	Around 5-6 high-risk townships in each of 300 counties73	198974	all counties: 6.1 (95% CIs not reported)75

Though these surveys provide some indication of baseline mortality rates in areas targeted for SIAs, they may differ from those areas in a few different ways:

• We do not know how the MNTE Initiative selected countries for surveys. Nigeria, Afghanistan and Niger, for instance, might have higher mortality rates than the areas typically targeted for SIAs, because those countries have among the highest under-5 mortality rates in the world.76
• While the surveys in Afghanistan, Lao People's Democratic Republic and China seem to have selected particularly high-risk areas to sample, the survey in Niger aimed for a nationally representative sample of the population and the survey in Nigeria aimed for a sample representative of Kano state, so the mortality rate in those countries may reflect both low and high risk areas.77
• All of the surveys occurred between 2001 and 2006 with the exception of the survey in China, which occurred in 1989.78 Mortality rates may have declined in the past several years.
• We do not know whether any SIAs occurred before the survey, so the areas surveyed may have lower mortality rates than the areas typically targeted for SIAs if previous SIAs had reduced the mortality rate.

National neonatal tetanus mortality rates

A relatively high national neonatal tetanus mortality rate suggests that there are a number of areas within the country that would benefit from SIAs. For instance, a large, nationally representative survey of India found a neonatal tetanus mortality rate of 1.2 per 1,000 live births, which exceeds the threshold for elimination.79

Relating clean delivery and immunization coverage rates to tetanus mortality

In areas where neonatal tetanus surveillance is poor, which includes most areas with high burden, the MNTE Initiative defines a high-risk area eligible for SIAs as an area where the reported clean delivery rate ("Delivery by a physician, nurse, or midwife, or as defined by national policy") is less than 70% and coverage with two doses of TT vaccine in pregnant women and women of childbearing age is less than 80%.80 We have not seen baseline surveys that could confirm whether the MNTE Initiative follows these guidelines in practice.

Areas that the MNTE Initiative defines as high-risk based on these guidelines may still have neonatal tetanus mortality rates below 1 death per 1,000 live births. For example, the MNTE Initiative conducted a validation survey in 3 districts in Nepal in 2005, which found a neonatal tetanus mortality rate of less than 1 death per 1,000 live births, a coverage rate of 2 doses of TT vaccine among women of childbearing age of 65% and a percentage of births assisted by health workers of 27%.81

The MNTE Initiative’s assessment and validation surveys suggest that areas deemed high-risk with a coverage rate of at least 2 doses of TT vaccine among mothers of around 50% will generally have a neonatal tetanus mortality rate greater than or equal to 1 death per 1,000 live births.82

Do SIAs cover their target population?

We have limited information indicating that SIAs achieve high coverage rates in their target population.

SIA coverage rates cited in MNTE Initiative reports

Some of the MNTE Initiative reports cite coverage rates achieved by SIAs.83 Based on the data in the reports, SIAs have a median coverage rate (the proportion of women targeted in a round that receive a dose) of 90% with an interquartile range of 79% to 94%.84 The reports provide little information on the source of this data.85 We believe the data comes from administrative reports, i.e. reports of vaccinations by the health workers involved in campaigns, which may overestimate coverage.86 We would also guess that SIAs discussed in MNTE reports achieve higher coverage rates than the typical SIA, because most of the MNTE reports focus on countries that have requested validation.87

Population coverage rates measured in validation and assessment surveys

The median proportion of mothers with at least 2 doses of TT vaccine from validation and assessment surveys is 75% and the median proportion of mothers with at least 3 doses is 37%.88 Assuming that coverage comes only from SIAs, this seems to imply a coverage rate achieved by SIAs roughly equivalent to reaching 56% of the target population with 3 doses, which is much lower than the SIA coverage rates discussed above.89 However, this approach may underestimate true coverage if the surveys took place in areas where a particularly low performing SIA had occurred.90 On the other hand, it may overestimate coverage if women received immunizations from multiple SIAs and antenatal clinics (for instance, women may have received immunizations from multiple SIAs each with poor coverage). It may also overestimate coverage if the surveys include doses that women received in infancy.

Do SIAs successfully immunize the population reached?

Even if SIAs reach their target population, they may still fail to successfully immunize that population. We have seen limited information to indicate that SIAs successfully immunize the target population.

Possible barriers to successful immunization include:

• An unreliable cold chain: The TT vaccine can lose its efficacy if not maintained at the appropriate temperature.91 A report assessing progress towards elimination in Nigeria in 2006 speculates that interruptions in the cold-chain may explain the high neonatal tetanus mortality despite high immunization coverage in Kano State.92
• Suboptimal timing of doses: The time between doses can influence the effectiveness of the vaccine.93 We have little information on the timing of rounds for SIAs in practice.
• Poor quality control in the manufacturing of vaccines: A study of TT immunization in Bangladesh estimated an efficacy of 45% due to sub-potent vaccine lots.94 Dietz et al. 1996 also found that “15 lots from eight manufacturers in seven countries had potency values below WHO requirements” out of a “total of 80 TT lots from 21 manufacturers in 14 of the 22 NT-reporting countries” tested.95
• A high prevalence of diseases that reduce maternal response to immunization: Some evidence suggests that maternal malaria and HIV/AIDS may reduce immunization response.96 We have not examined these studies.

How much does the program cost?

We have limited information about the costs to implement this program. We suspect that a review of expenditures by UNICEF's MNTE Initiative or another organization implementing this program would produce the most relevant estimates of cost.

We are aware of 2 studies estimating the cost of SIAs in high-risk areas to prevent neonatal tetanus. Note that we have not undertaken a thorough literature review to identify other estimates of cost, nor have we carefully vetted Griffiths et al. 2004 and Berman et al. 1991.

• Griffiths et al. 2004 examined SIAs conducted in a high-risk district in Pakistan from 2001 to 2003.97 They estimated the cost of the program at $1.19 (about $1.50 in 2013 dollars) per woman vaccinated with three doses.98
• Berman et al. 1991 examined SIAs conducted in the Aceh province of Indonesia in 1985.99 They estimated the cost of the program at $1.59 (about $3.40 in 2013 dollars) per woman vaccinated with two doses.100

Cost-effectiveness

The cost-effectiveness of this program is highly sensitive to several inputs about which we have limited information (discussed above). We would seek additional information about these inputs from any organization we consider for a recommendation.

We have completed a rough cost-effectiveness analysis of various scenarios for this program: our rough cost-effectiveness model (xlsx).101 We believe there are plausible scenarios where this program could be competitive with or significantly more cost-effective than our top charities.

Possible negative impacts

• Mild local reactions to the vaccine are relatively common, but severe adverse reactions appear extremely rare.102
• Needles may theoretically be used on individuals infected with a disease and then reused on others, which can contribute to disease transmission.103 We are not aware of any concerns related to needle safety particular to SIAs to prevent tetanus.
• Relying too heavily on SIAs may hamper local efforts to improve health systems and distort local priorities.104 Donor investment in SIAs could cause countries to reduce their investment in routine immunization systems.105 We are unsure of the conditions under which donor-funded, disease-specific programs weaken health systems.106

Our process

We relied on Demicheli, Barale and Rivetti 2013 for its review of the evidence on maternal tetanus immunization. We relied on Maternal and Neonatal Tetanus (MNT) related publications and useful links for its list of MNTE Initiative reports on validation and assessment surveys. We relied on those reports to try to answer whether SIAs target populations at risk, whether SIAs reach their target population and whether the vaccines SIAs deliver remain effective. Roper, Vandelaer, and Gasse 2007 provided a general discussion of maternal and neonatal tetanus. Stroh and Birmingham 2002 described the WHO’s recommendations on the methodology used to assess coverage rates and to determine whether the neonatal mortality rate in an area exceeds 1 per 1,000. We relied on Achieving and Sustaining Maternal and Neonatal Tetanus Elimination: Strategic Plan 2012-2015 for its discussion of the MNTE Initiative and the strategy underlying SIAs. We had a conversation with Dr. Neal Halsey, Professor of International Health and Director of the Institute for Vaccine Safety at Johns Hopkins Bloomberg School of Public Health, about the relationship between different immunization schedules and duration of immunity.

Questions for a charity

While not an exhaustive list, we would ask the following questions of any charity applying for a recommendation based on its work implementing SIAs to combat maternal and neonatal tetanus:

• What is the evidence that there is a high rate of neonatal tetanus mortality among the populations you serve?
• What is the coverage of DTP vaccine in the populations you serve?
• What is the coverage of TT vaccine in the populations you serve?
• What is the evidence that the SIAs you implement reach a high proportion of the women they target?
• How much do the SIAs you implement cost you and other organizations and governments?
• What is the evidence that the SIAs you implement successfully immunize the population they cover?
• How much time passes between rounds of the SIAs you implement?
• Is it possible to fund SIAs exclusively?107

Questions for further investigation

• What is the relationship between different immunization schedules and duration of immunity?108
• What is the relationship between the timing of rounds and the effectiveness of SIAs?109
• Does a high prevalence of certain diseases reduce maternal response to immunization?110
• What evidence does Koenig et al. 1998, an observational study following up on Black, Huber and Curlin 1980 over around a 13 year period after immunization, provide about the duration of immunity?111
• We could refine our cost-effectiveness model by more precisely modeling the effect of the number of TT doses received, accounting for the fact that women who are not already immunized before an SIA are likely harder to reach and thus less likely to be immunized during the SIA, accounting for tetanus morbidity, or modeling the age structure and age-specific fertility rates of the population. What effect would these refinements have on our cost-effectiveness estimate?
• How strong is the evidence for the effectiveness of other interventions to combat tetanus such as promotion of clean birth practices, immunization at fixed sites (e.g. antenatal clinics), and training of skilled birth attendants?

Sources

Document	Source
Achieving and Sustaining Maternal and Neonatal Tetanus Elimination: Strategic Plan 2012-2015	Source (archive)
Atkinson, Wolfe and Hamborsky eds. 2012	Source (archive)
Baltazar and Sarol 1994	Source (archive)
Berman et al. 1991	Source (archive)
Black, Huber and Curlin 1980	Source (archive)
Blencowe et al. 2010	Source (archive)
Blencowe et al. 2011	Source (archive)
Borrow, Balmer and Roper 2006	Source (archive)
Chai et al. 2004	Source (archive)
Chongsuvivatwong et al. 1993	Source (archive)
CPI Inflation Calculator	Source
Demicheli, Barale and Rivetti 2013	Source (archive)
Dietz et al. 1996	Source (archive)
Elimination of Maternal and Neonatal Tetanus, UNICEF	Source (archive)
GiveWell’s non-verbatim summary of a conversation with Dr. Neal Halsey, Professor of International Health and Director of the Institute for Vaccine Safety, Johns Hopkins Bloomberg School of Public Health, January 13, 2015	Source
Griffiths et al. 2004	Source (archive)
Guilfoile and Babcock 2008	Source (archive)
Gupta and Keyl 1998	Source (archive)
Hlady et al. 1992	Source (archive)
Immunization Country Reports, Childinfo	Source (archive)
Jamison et al. eds. 2006	Source (archive)
Koenig et al. 1998	Source (archive)
Lim et al. 2008	Source (archive)
Lu et al. 2010	Source (archive)
Maternal and Neonatal Tetanus (MNT) related publications and useful links	Source (archive)
Million deaths study 2010	Source (archive)
Msuya 2004	Source (archive)
Newell et al. 1966	Source (archive)
Notes from a conversation with Rownak Khan on June 22, 2012	Source
Rahman et al. 1982	Source (archive)
Report on progress towards MNT elimination, Afghanistan (Badakshan, Kandahar, Kunduz provinces), 2004	Source (archive)
Report on progress towards MNT elimination, China, 1993	Source (archive)
Report on progress towards MNT elimination, Global update 1993	Source (archive)
Report on progress towards MNT elimination, Global update 1989-1993	Source (archive)
Report on validation of MNTE elimination in countries, India, 2 States, 2008	Source (archive)
Report on progress towards MNT elimination, India (Rajastan), 2003	Source (archive)
Report on progress towards MNT elimination, Indonesia, 2003	Source (archive)
Report on progress towards MNT elimination, Lao People's Democratic Republic, 2002	Source (archive)
Report on progress towards MNT elimination, Madagascar, 2009	Source (archive)
Report on progress towards MNT elimination, Mali, 2007	Source (archive)
Report on progress towards MNT elimination, Niger, 2002	Source (archive)
Report on progress towards MNT elimination, Nigeria (Kano State), 2006	Source (archive)
Report on progress towards MNT elimination, Tanzania, 2009	Source (archive)
Report on progress towards MNT elimination, Thailand, 1993	Source (archive)
Report on validation of MNTE elimination in countries, Bangladesh, 2008	Source (archive)
Report on validation of MNTE elimination in countries, Burundi and the Comoros, 2009	Source (archive)
Report on validation of MNTE elimination in countries, Congo (the), 2009	Source (archive)
Report on validation of MNTE elimination in countries, Egypt, 2007	Source (archive)
Report on validation of MNTE elimination in countries, Eritrea, 2004	Source (archive)
Report on validation of MNTE elimination in countries, Indonesia, 2010	Source (archive)
Report on validation of MNTE elimination in countries, Mozambique, 2010	Source (archive)
Report on validation of MNTE elimination in countries, Myanmar, 2010	Source (archive)
Report on validation of MNTE elimination in countries, Nepal, 2006	Source (archive)
Report on validation of MNTE elimination in countries, Togo, 2006	Source (archive)
Report on validation of MNTE elimination in countries, Turkey, 2009	Source (archive)
Report on validation of MNTE elimination in countries, Viet Nam, 2006	Source (archive)
Roper, Vandelaer, and Gasse 2007	Source (archive)
Snow et al. 1992	Source (archive)
Stroh and Birmingham 2002	Source (archive)
Tetanus deaths by age, GBD Compare	Source (archive)
Tetanus deaths by age, female, GBD Compare	Source (archive)
Tetanus deaths over time, GBD Compare	Source (archive)
Schofield, Tucker and Westbrook 1961	Source (archive)
UNICEF MNTE Initiative Budget, June 2014	Unpublished
Wang et al. 2011	Source (archive)
Wang et al. 2014	Source (archive)
WHO 2006	Source (archive)
World Bank Data, Under-5 mortality	Source (archive)
Yusuf et al. 1991	Source (archive)

• 1

  “As of September 2019, 47 out of 59 countries identified as high-risk for maternal and newborn tetanus had fully eliminated the disease.” (UNICEF Maternal and Newborn Health, accessed 1/2/2020)

• 2

  “Through The Eliminate Project, Kiwanis International and UNICEF have joined forces to eliminate maternal and neonatal tetanus—a deadly disease that steals the lives of nearly 31,000 innocent babies and a significant number of women each year...we reached our fundraising goal, raising US$110 million in cash and pledges since 2010.” (The Eliminate Project, accessed 1/2/2020)

• 3

  Angola, Central African Republic, Guinea, Mali, Nigeria, Somalia, South Sudan, Sudan, Afghanistan, Pakistan, Yemen, Papua New Guinea (Kiwanis: Maternal and Neonatal Tetanus Update, accessed 1/2/2020)

• 4

  “UNICEF has helped to successfully eliminate MNT in many countries. But in 12 countries it still strikes babies and mothers who have little or no access to health care — either because they are poor, live in remote areas or are caught in humanitarian emergencies. More funds and resources are needed to reach all babies and mothers at risk.” (Kiwanis: Maternal and Neonatal Tetanus, accessed 1/2/2020)

• 5

  "Tetanus is characterised by muscle rigidity and painful muscle spasms caused by tetanus toxin’s blockade of inhibitory neurons that normally oppose and modulate the action of excitatory motor neurons." Roper, Vandelaer, and Gasse 2007, Pg. 1949

• 6
  • "Tetanus is caused by a neurotoxin produced by Clostridium tetani, a gram-positive, obligate anaerobic rod-shaped bacterium that forms spores. C tetani spores occur worldwide as constituents of soil and in the gastrointestinal tracts of animals (including human beings), and can contaminate many surfaces and substances." Roper, Vandelaer, and Gasse 2007, Pgs. 1947-1948
  • "The spores are also relatively resistant to phenol and other chemical agents. The spores are widely distributed in soil and in the intestines and feces of horses, sheep, cattle, dogs, cats, rats, guinea pigs, and chickens. Manure-treated soil may contain large numbers of spores. In agricultural areas, a significant number of human adults may harbor the organism. The spores can also be found on skin surfaces and in contaminated heroin." Atkinson, Wolfe and Hamborsky eds. 2012, Pg. 291
  • "C. tetani usually enters the body through a wound. In the presence of anaerobic (low oxygen) conditions, the spores germinate. Toxins are produced and disseminated via blood and lymphatics. Toxins act at several sites within the central nervous system, including peripheral motor end plates, spinal cord, and brain, and in the sympathetic nervous system. The typical clinical manifestations of tetanus are caused when tetanus toxin interferes with release of neurotransmitters, blocking inhibitor impulses. This leads to unopposed muscle contraction and spasm. Seizures may occur, and the autonomic nervous system may also be affected." Atkinson, Wolfe and Hamborsky eds. 2012, Pg. 292

• 7
  • "Increased tetanus toxoid vaccination coverage and hygienic intrapartum and postnatal practices, particularly cord care, are important contributing factors [14,15]. In addition to variation in immunization coverage, intrapartum and postnatal practices may explain much of the local variation in incidence of tetanus [16-18]. Clean birth practices have been associated with dramatic reductions in the incidence of neonatal tetanus in the absence of immunization, for example in industrialized countries where tetanus was virtually eliminated before the vaccine was introduced and. in China, training of traditional birth attendants (TBAs) and providing them with a ‘clean birth kit’ in the 1950s led to a reduction in neonatal tetanus rates from 32/1000 in 1948 to 2/1000 in 1961 [19]." Blencowe et al. 2011, Pg. 2
  • "Factors associated with unsafe procedures
    • Deliveries or medical procedures done outside health-care facilities*
    • Birth attendants without medical training*
    • Unclean hands* and instruments*
    • Dirt,* straw,* or other unclean materials as delivery surface
    • Animals kept inside or adjacent to home (for home deliveries)
    • Animal dung used for fuel
    • Traditional substances used during labour, delivery, or
      abortion (ie, cow ghee* and other animal or vegetable oils,* juices or herbs)
    • Traditional substances used for umbilical cord care (ie, cow
      dung,* rat faeces, cow ghee, other oils or juices, herbs, ash,* surma,* soil, sand)
    • Neonates swaddled in animal dung* or soil
    • Traditional neonatal surgeries (ie, circumcision, ritual
      scarification, ear piercing, uvulectomy)

    Immunisation-related factors

    • Absent or incomplete immunisation with tetanus toxoid

    Factors associated with unsafe procedures or incomplete immunisation, or both

    • Poverty*
    • Absent or poor maternal* or paternal education,* or both
    • Poor antenatal-care attendance*
    • Young maternal age or first pregnancy,* or both
    • Cultural constraints to women’s movements and contacts

    Other factors

    • Death of a previous child in a family from neonatal tetanus (predictive of subsequent cases)*
    • Male sex (increased risk of neonatal tetanus)‡

    Specific factors can be related to unsafe delivery, abortion, or cord-care practices, or to inadequate immunisation with tetanus toxoid, or both. *Identified as independent risk factors for neonatal tetanus by multivariate analysis. References for studies using multivariate analysis to identify independent risk factors. ‡Inconsistent findings in community-based and hospital-based studies; unclear if related to differential cord care, maternal recall, or medical-care seeking for males, or because of a biological difference between boys and girls." Panel 1: Risk factors associated with neonatal and maternal tetanus, Roper, Vandelaer, and Gasse 2007, Pg. 1952.

• 8

  "The risk of death is highest for very young and very old patients. Neonatal tetanus mortality approached 100% in community-based surveys in the 1980s, but is now 10–60% with hospital care." Roper, Vandelaer, and Gasse 2007, Pg. 1950

• 9
  • Decline in tetanus deaths over time: Tetanus deaths over time, GBD Compare
  • Most tetanus deaths occur in the neonatal period: Tetanus deaths by age, GBD Compare

• 10

  "Maternal and neonatal tetanus cases are clustered in poor, remote, and disenfranchised communities where unhygienic obstetric and postnatal practices prevail, and access to maternal tetanus toxoid immunisation is poor. Differences in neonatal tetanus incidence and mortality of at least an order of magnitude have been identified between regions and countries, and between urban and rural areas within countries. In industrialised countries, neonatal tetanus ceased to be a substantial problem by the mid-20th century: once tetanus toxoid vaccination became widespread, neonatal tetanus disappeared." Roper, Vandelaer, and Gasse 2007, Pg. 1952

• 11

  "The only reliable immunity against tetanus is that induced by vaccination with tetanus toxoid. Tetanus toxoid vaccine is one of the most effective, safe, stable, and inexpensive vaccines ever developed, and can be given safely during pregnancy and to immunocompromised individuals. It is available as single-antigen vaccine and in many multiple-antigen preparations. When handled and given properly, it provides highly protective, longlasting immunity against tetanus." Roper, Vandelaer, and Gasse 2007, Pgs. 1950-1951
  

• 12

  See Are the babies born to vaccinated women protected?

• 13

  "In recognition of the substantial burden of neonatal tetanus in developing countries, the 1989 World Health Assembly (WHA) adopted a resolution to eliminate neonatal tetanus by 1995, through the increased availability of tetanus toxoid and clean deliveries, and improved surveillance." Roper, Vandelaer, and Gasse 2007, Pg. 1954

• 14
  • "The elimination of neonatal tetanus was defined as fewer than 1 case per 1000 livebirths in every district. This definition also has been adopted as a proxy for the elimination of maternal tetanus." Roper, Vandelaer, and Gasse 2007, Pg. 1954
  • “Since tetanus spores cannot be removed from the environment, sustaining elimination will require improvements to presently inadequate immunisation and health-service infrastructures, and universal access to those services.” Roper, Vandelaer, and Gasse 2007, Pg. 1947
  • "C tetani spores occur worldwide as constituents of soil and in the gastrointestinal tracts of animals (including human beings), and can contaminate many surfaces and substances." Roper, Vandelaer, and Gasse 2007, Pgs. 1947-1948

• 15

  "In areas with reasonably well-developed health services, improvements in tetanus toxoid vaccination coverage were fairly easy to achieve, leading to a 25% reduction in worldwide deaths from neonatal tetanus by 1992. However, elimination activities continued to miss communities with a high burden of neonatal tetanus and poor access to routine health services. The high-risk approach was introduced to address this shortcoming (panel 2, figures 5 and 6)…As cost effective as tetanus toxoid campaigns are, in the absence of external funding to support the supplemental activities, the implementation of the high-risk approach was weak in the countries most in need. As a result, the 1995 target date for global elimination of neonatal tetanus was not met." Roper, Vandelaer, and Gasse 2007, Pg. 1954
  

• 16

  “Due to the nature of the disease, tetanus cannot be eradicated and so the goal is to eliminate MNT as a public health problem as called for by the 42nd World Health Assembly in 1989 and endorsed by the World Summit for Children in 1990 and later reinforced by the re-launch of the maternal and neonatal tetanus elimination (MNTE) initiative by the World Health Organization (WHO), United Nations Children’s Fund (UNICEF) and the United Nations Population Fund (UNFPA) in 1999.” Achieving and Sustaining Maternal and Neonatal Tetanus Elimination: Strategic Plan 2012-2015, Pg. 3

• 17

  "Once a country concludes that it has met the definition of MNTE, a request is made to WHO headquarters (HQ) to conduct a formal assessment, and usually a community-based neonatal mortality survey is performed in the district at highest risk for MNT to validate the elimination status. If the survey shows that the highest-risk district has met the definition of MNTE of less than 1 NT case per 1,000 live births, it is assumed that all districts at lower risk also have achieved elimination, as has the country as a whole." Achieving and Sustaining Maternal and Neonatal Tetanus Elimination: Strategic Plan 2012-2015, Pg. 17

• 18
  • "During previous assessments a procedure was developed for identifying districts at highest risk for NT. It is based on three steps: 1) consolidating core and surrogate indicators on every district in a spreadsheet (Annex 1); 2) developing a short list of districts potentially at highest risk by ranking the districts according to the values of the indicators; 3) selecting the districts to be surveyed from the short list on the basis of public health judgement and logistical and administrative considerations." Stroh and Birmingham 2002, Pg. 8
  • As an example of how the protocol was implemented: "Only districts where more than 65% of the population lived in rural areas and TT3 coverage in SIAs was ≤50% were selected. Four districts met these criteria: Abeldara, Tin Essako, Gao and Nioro. The first 2 were ruled out because their population was too small. Gao was selected over Nioro because of its greater rural population and lower immunization coverage (especially TT2) in SIAs...The district where the neonatal tetanus rate was estimated to be among the highest in the country was thus deliberately chosen to prove that, if neonatal tetanus had been eliminated there, the disease had probably also been eliminated in districts with better indicators." Report on progress towards MNT elimination, Mali, 2007, Pg. 310

• 19
  • "In order to evaluate the neonatal tetanus mortality rate (NTMR) in a community, the survey method described in this document uses the principles of lot quality assurance sampling to judge whether the NTMR has been reduced to less than 1/1000 live births during a recent 12-month interval. This method is appropriate to use at the final stage of MNT elimination when there is evidence to suggest that NT is reduced to less than 1/1000 live births and is only occurring sporadically, i.e. not clustering. Using this method in combination with cluster survey sampling, data are collected on clusters of births in order to reduce costs." Stroh and Birmingham 2002, Pg. 12
  • "The quality assurance sampling approach allows a population to be classified as acceptable or not acceptable in respect of a particular attribute. The decision to accept or not accept is based on the mathematical probabilities of obtaining fewer than a specified number of units with a particular attribute in a random sample of predetermined size drawn from the population to be classified. The major advantage of using a quality assurance sampling design is that smaller sample sizes are needed than in conventional sample surveys in which point estimates with designated precisions are desired." Stroh and Birmingham 2002, Pg. 29
  • "In summary, this method was considered the most practical for assessing whether MNT elimination has been achieved; if districts at highest risk are surveyed and 'pass', it is reasonable to assume that other districts (at lower risk) have also achieved MNT elimination. Community-based conventional surveys are impractical for this purpose because they would require very large sample sizes (e.g. tens of thousands of live births) to measure a low rate of NT mortality." Stroh and Birmingham 2002, Pg. 30

• 20
  • "If the survey shows that the highest-risk district has met the definition of MNTE of less than 1 NT case per 1,000 live births, it is assumed that all districts at lower risk also have achieved elimination, as has the country as a whole." Achieving and Sustaining Maternal and Neonatal Tetanus Elimination: Strategic Plan 2012-2015, Pg. 17
  • "In summary, this method was considered the most practical for assessing whether MNT elimination has been achieved; if districts at highest risk are surveyed and 'pass', it is reasonable to assume that other districts (at lower risk) have also achieved MNT elimination. Community-based conventional surveys are impractical for this purpose because they would require very large sample sizes (e.g. tens of thousands of live births) to measure a low rate of NT mortality." Stroh and Birmingham 2002, Pg. 30

• 21

  "The World Health Assembly first called for elimination of neonatal tetanus in 1989. In 1999, the goal was expanded to include elimination of the maternal tetanus. At that time, there were 57 countries that had still not eliminated MNT. The figure today stands at 59 with inclusion of Timor Leste in 2002 and South Sudan in 2011. The goal of the initiative is to eliminate maternal and neonatal tetanus (MNT) through focus on the TT-SIAs. As of December 2013, 34 of these countries had achieved MNT elimination leaving 25 countries that still have not eliminated the disease." Elimination of Maternal and Neonatal Tetanus, UNICEF

• 22

  "High-Risk Area Approach to MNTE 1. Identification of high-risk districts/areas in a country, based on a thorough data review. 2. Implementation of three rounds of TT-SIAs in all of the high-risk districts to immunize all women of reproductive age, irrespective of their previous vaccination status. 3. Minimum of four weeks interval between the first and second rounds and a minimum of six months between the second and third rounds. 4. Support local initiatives to promote clean delivery and clean cord care practices." Achieving and Sustaining Maternal and Neonatal Tetanus Elimination: Strategic Plan 2012-2015, Pg. 11

• 23

  "3. Immunization of women of reproductive age with TT or Td vaccine, through SIAs: MNT is a disease of underserved populations and special groups such as nomads and displaced persons. In such areas, major infrastructural changes, which are not easy to implement, are needed to provide preventive and curative services for NT. Therefore, such areas and populations require special strategies to overcome system barriers. These areas are identified through data reviews that look at a range of indicators, including immunization coverage, surveillance data, maternal health indicators, health systems indicators, etc. Once the high-risk districts are identified, the supplemental immunization is implemented through i. Stand-alone TT/Td-SIAs (single intervention campaigns) ii. Integration of TT/Td-SIAs with other interventions such as measles or polio vaccination, vitamin A supplementation, mosquito net distribution, commodities distribution, etc. iii. Child Health Days (CHDs) or similar periodic multi-intervention campaigns" Achieving and Sustaining Maternal and Neonatal Tetanus Elimination: Strategic Plan 2012-2015, Pg. 11

• 24

  "Since its inception in 1990 the neonatal tetanus elimination initiative has stressed focus on communities traditionally missed by routine immunisation and maternal and child health services, because of geographical and sociocultural barriers to health-service access…After high-risk areas have been selected, supplemental immunisation activities are organised, targeting all women of childbearing age (usually 15–45 years) with three doses of tetanus toxoid." Panel 2: The high-risk approach, Roper, Vandelaer, and Gasse 2007, Pg. 1954

• 25
  • "Budget: Achieving MNT elimination mainly through TT-SIAs targeting more than 100 million women who live in areas still at risk for MNT will cost an estimated USD 227 million. A funding gap of approximately USD 102 million has to be met between 2012 and 2015 in order to achieve global elimination which is programmatically feasible by 2015. Sustaining MNTE is part of the existing immunization budget that may be augmented by the cost to introduce and deliver booster doses and inclusion of maternal health budgets." Achieving and Sustaining Maternal and Neonatal Tetanus Elimination: Strategic Plan 2012-2015, Pg. 19
  • "1. Delivery by skilled birth attendants to ensure clean delivery practices: Safe birth is a basic right of mothers and newborns. Clean birthing practices and umbilical cord care can substantially reduce maternal and neonatal mortality and morbidity from infectious causes, including tetanus. Changes in both traditional practices, training of high-quality skilled birth attendants and development of functional obstetric facilities are required. MNT can be prevented by ensuring that all deliveries are assisted by skilled health attendants – whether at health facilities or during home delivery – who can ensure clean delivery practices. It is also important to discourage use of harmful cord care practices through information and education of communities.
    2. Immunization of women during pregnancy (at fixed sites or through outreach) with TT or tetanus diphtheria toxoid (Td) vaccines: Tetanus vaccination actually begins in infancy, when children receive at least three doses of diphtheria, pertussis and tetanus (DPT) vaccine. However, the duration of protection from infant immunization wanes and booster doses in school-age children required to maintain protection are not available in most developing countries, especially in areas most affected by NT. Therefore, immunizing women during pregnancy, and following review of their vaccination history and current protection, is recommended to provide protection against tetanus. This can be done during antenatal care (ANC) visits or immunization sessions at fixed sites or during outreach.
    3. Immunization of women of reproductive age with TT or Td vaccine, through SIAs: MNT is a disease of underserved populations and special groups such as nomads and displaced persons. In such areas, major infrastructural changes, which are not easy to implement, are needed to provide preventive and curative services for NT. Therefore, such areas and populations require special strategies to overcome system barriers. These areas are identified through data reviews that look at a range of indicators, including immunization coverage, surveillance data, maternal health indicators, health systems indicators, etc. Once the high-risk districts are identified, the supplemental immunization is implemented through i. Stand-alone TT/Td-SIAs (single intervention campaigns) ii. Integration of TT/Td-SIAs with other interventions such as measles or polio vaccination, vitamin A supplementation, mosquito net distribution, commodities distribution, etc. iii. Child Health Days (CHDs) or similar periodic multi-intervention campaigns.
    4. Surveillance for NT: to identify where cases still occur and why, helping direct programmatic efforts and monitor MNTE status." Achieving and Sustaining Maternal and Neonatal Tetanus Elimination: Strategic Plan 2012-2015, Pgs. 10-11

• 26

  "Tetanus toxoid vaccine is one of the most effective, safe, stable, and inexpensive vaccines ever developed, and can be given safely during pregnancy and to immunocompromised individuals. It is available as single-antigen vaccine and in many multiple-antigen preparations. When handled and given properly, it provides highly protective, longlasting immunity against tetanus." Roper, Vandelaer, and Gasse 2007, Pgs. 1950-1951
  

• 27

  "The advent of the vaccine resulted in further reduction in high-income countries, and also opened opportunities for progress in low-income settings. The vaccine is an inactivated toxin (toxoid) that was first produced in 1924. It became commercially available in 1938 and was successfully used extensively during the Second World War. In the late 1940s, it was combined with diphtheria and pertussis vaccines to produce the DTP triple vaccine used in many childhood immunization programmes." Blencowe et al. 2010, Pg. i103
  

• 28
  • "In 1926 Ramon and Zoller described the use of formaldehyde-inactivated tetanus toxoid as a vaccine for humans. Essentially, nearly the same formulation is used today. Initially, one of their colleagues, Lafaille tested the vaccine preparation on himself. Noting no ill effects on Lafaille, Ramon and Zoller then vaccinated 100 people with the tetanus toxoid. No harmful effects were noted in the people who were vaccinated. They found that a single dose of the vaccine did not result in the production of much antibody. This was based on the observation that 1 cc of blood serum drawn from test subjects did not contain enough antibody to neutralize the small amount of toxin required to kill a single guinea pig. However, those subjects who received two vaccinations had sufficient antibody in 1 cc of blood serum to neutralize an amount of tetanus toxin that would kill 1,000 to 3,000 guinea pigs. This was a much higher level of antibody production than Eisler had previously observed with his vaccine experiments. This information helped establish that multiple vaccinations are needed for people to be protected against tetanus. In addition Ramon and Zoller showed that this vaccine could be mixed with another vaccine to confer resistance to multiple pathogens with a single injection. This formed the basis for the modern DPT shot, which protects against diphtheria, pertussis, and tetanus, disease caused by three different species of bacteria." Guilfoile and Babcock 2008, Pgs. 33-34
  • "Efficacy of the toxoid has never been studied in a vaccine trial. It can be inferred from protective antitoxin levels that a complete tetanus toxoid series has a clinical efficacy of virtually 100%; cases of tetanus occurring in fully immunized persons whose last dose was within the last 10 years are extremely rare." Atkinson, Wolfe and Hamborsky eds. 2012, Pg. 297

• 29
  • "We included two trials involving 95,704 immunised individuals (including women of childbearing age, pregnant women and children aged one to 14 years) and 10,560 infants born to them...One trial used individual randomisation (Newell 1966) while the other did not report details of randomisation methods and was included as a quasi-randomised trial (Black 1980)." Demicheli, Barale and Rivetti 2013, Pg. 7
  • Quality of the evidence: "In conclusion, this review shows that vaccination with tetanus toxoid is effective in preventing neonatal tetanus cases and deaths specifically caused by neonatal tetanus. Even if the evidence is derived mainly from a single study this appears to be solid and consistent with the findings of all other comparative studies presently available.” Demicheli, Barale and Rivetti 2013, Pg. 11
  • Overall completeness and applicability of evidence: “Therefore, the reasons for the low performance presently achieved by the vaccination campaigns should be sought outside the field of vaccine efficacy and are probably related to organisational and quality issues. Our review did not find evidence on the main factors that can have negative influence on the impact of the immunisation practice and that may justify the present low level of performance of the campaign (Dietz 1996; WHO 1999).” Demicheli, Barale and Rivetti 2013, Pg. 11
  • Authors’ conclusions: "Available evidence supports the implementation of immunisation practices on women of childbearing age or pregnant women in communities with similar, or higher, levels of risk of neonatal tetanus, to the two study sites. More information is needed on possible interference of vaccination by malaria chemoprophylaxis on the roles of malnutrition and vitamin A deficiency, and on the quality of tetanus toxoid production and storage.” Demicheli, Barale and Rivetti 2013, Pg. 2
  • Summary of main results: "The effectiveness of vaccination in preventing deaths from neonatal tetanus appears to be high when two or more doses are administered. The vaccination does not exert effects on causes of death other than tetanus. The two studies apparently show differences in the estimates of effect, but these differences are understand- able when considering that the study showing lower effectiveness (Black 1980) was assessing a different intervention (vaccination with only one or two doses) and a less specific outcome (all causes neonatal deaths occurring four to 14 days from birth). In the in- terpretation of the results it must be also considered that one study (Newell 1966) had a third arm containing participants who refused vaccination and that data from this arm were not included in the analysis. Because of the limited number of eligible experi- mental studies included in this review, we decided to carry out an extended search in order to retrieve all the available comparative studies on this topic. The same databases were explored in order to identify cohort, case-control studies and other non-randomised study designs. Seven further studies were identified (four surveys comparing the disease incidence before and after the introduction of the immunisation campaign, two case-control studies and a co- hort study). The characteristics of these studies are summarised in the additional Table 2. Altogether 37,352 births were surveyed by the prospective studies and 552 participants were included in the case-control studies. All the studies but one confirmed the existence of a significant protective effect of an immunisation course of at least two doses of tetanus toxoid on the incidence of neonatal tetanus." Demicheli, Barale and Rivetti 2013, Pg. 11
  • Plain language summary: ““Neonatal tetanus is an infection causing rigidity, muscle spasm and often death in newborn babies. It is quite common in income-poor countries and comes from insufficient protection being passed from mother to baby in utero together with infection entering into the baby through the umbilical cord stump.

    This review assessed the effects of vaccinating women of childbearing age and included two studies (10,560 infants). One study assessed the effects of aluminium phosphate adsorbed tetanus toxoid against polyvalent influenza and the other study assessed the effects of adsorbed tetanus-diphtheria toxoid against cholera toxoid. Vaccinating childbearing women against tetanus rather than influenza or cholera appears to decrease the incidence of tetanus in newborn babies. Vaccinating women of childbearing age showed fewer cases of neonatal tetanus when two or three doses were used, but no potential adverse effects were assessed in either study.

    Caution needs to be taken in interpreting these results, because the results are based on outcomes of the infants born to only a subset of those women randomised. Administrative and operational aspects also need to be of good quality for vaccination programmes to be effective." Demicheli, Barale and Rivetti 2013, Pg. 2

  • "Immunization of non-pregnant women in rural Bangladesh with two doses of aluminium-adsorbed tetanus-diphtheria toxoids reduced neonatal mortality by one-third during a period of 9-32 months after vaccination." Black, Huber and Curlin 1980, Pg. 927
  • "The study group was given 1-3 injections of I ml of an aluminium-phosphate-adsorbed tetanus toxoid more than 6 weeks apart, and the control group a similar number of injections of an influenza-virus vaccine. There was no statistically significant difference between those in the two groups given one injection. Those in the control group given 2 or 3 injections had a tetanus neonatorum death rate of 7.8 deaths per 100 births, and the corresponding subjects in the study group had none. This difference is unlikely to have occurred by chance." Newell et al. 1966, Pg. 863

• 30

  "To avoid these possibilities of error, it was decided that the trial would have to be a comparison, within one area, between two groups of infants whose mothers had been randomly allocated to a study and a control group." Newell et al. 1966, Pg. 864

• 31

  "In the trial with influenza vaccine as control (Newell 1966), 2776 women aged between 13 and 45 years were enrolled. They were randomised to receive three doses of one vaccine preparation. Of these, 1158 declined to receive any immunisation and their infants were not included in the analysis (n = 601). Also, 136 infants born to the immunised groups were not included in the analysis because they were born before mothers could receive the first dose of the vaccine. Overall, 1182 infants were included in the analysis. A total of 9823 births were considered from the two studies combined.” Demicheli, Barale and Rivetti 2013, Pgs. 7-8

• 32

  "…a double-blind field trial covering 1618 women was conducted between 1961 and 1966 in a rural area of Colombia with an estimated existing tetanus neonatorum death rate of 11.6 per 100 births." Newell et al. 1966, Pg. 863

• 33

  "One dose", Analysis 1.1, Demicheli, Barale and Rivetti 2013, Pg. 22

• 34
  • "Two or three doses", Analysis 1.1, Demicheli, Barale and Rivetti 2013, Pg. 22
  • Among women receiving an injection, about 32% received 1 dose, about 18% received 2 doses and about 50% received 3 doses. Table 1, Newell et al. 1966, Pg. 866.

• 35

  “Interval from injection to birth (months)”. Tables 5-6, Newell et al. 1966, Pg. 869

• 36

  Analysis 1.1, Demicheli, Barale and Rivetti 2013, Pg. 22

• 37
  • "One trial used individual randomisation (Newell 1966) while the other did not report details of randomisation methods and was included as a quasi-randomised trial (Black 1980)." Demicheli, Barale and Rivetti 2013, Pg. 7
  • "In 1974, a group of subjects was vaccinated with a cholera toxoid with a view to evaluating its efficacy. Concomitantly, aluminium phosphate-adsorbed tetanus and diphtheria (Td) toxoids were given to a randomly-assigned group of subjects whose served as controls." Black, Huber and Curlin 1980, Pg. 927

• 38

  "The trial that compared tetanus-diphtheria toxoid with cholera toxoid (Black 1980) included a total of 92,928 healthy women aged at least 15 years, and children aged one to 14 years, who were immunised with one or two doses of the vaccine preparations. Follow-up was performed on 8641 infants born from this group of women and began nine months after immunisation, to ensure that women pregnant at the time of vaccination had been excluded from the analysis.” Demicheli, Barale and Rivetti 2013, Pg. 7

• 39

  "Immunization of non-pregnant women in rural Bangladesh with two doses of aluminum-adsorbed tetanus-diphtheria toxoids reduced neonatal mortality by one-third during a period of 9-32 months after vaccination." Black, Huber and Curlin 1980, Pg. 927

• 40

  Treatment effect by doses (xlsx)

• 41

  Treatment effect by doses (xlsx)

• 42

  “Immunization of non-pregnant women in rural Bangladesh with two doses of aluminium-adsorbed tetanus-diphtheria toxoids reduced neonatal mortality by one-third during a period of 9-32 months after vaccination.” Black, Huber and Curlin 1980, Pg. 927

• 43
  • "We have used neonatal mortality, and more precisely mortality on days 4-14, as indicators of neonatal tetanus. Earlier studies of unimmunized populations indicate that in some countries tetanus is responsible for one-fourth to one-third of all neonatal deaths(1). Furthermore, deaths between days 4 and 14 after birth should be a good indicator of tetanus, since studies have demonstrated that over 90% of neonatal deaths caused by tetanus, but only 25% of neonatal deaths from other causes, would occur in this period (3)." Black, Huber and Curlin 1980, Pg. 929
  • Reduction in neonatal all-cause mortality was 32% (95% CI: 18%-44%) Analysis 2.1, Demicheli, Barale and Rivetti 2013, Pg. 25

• 44

  "In the trial with influenza vaccine as control (Newell 1966), 2776 women aged between 13 and 45 years were enrolled. They were randomised to receive three doses of one vaccine preparation.” Demicheli, Barale and Rivetti 2013, Pg. 7

• 45

  "In the trial with influenza vaccine as control (Newell 1966), 2776 women aged between 13 and 45 years were enrolled. They were randomised to receive three doses of one vaccine preparation. Of these, 1158 declined to receive any immunisation and their infants were not included in the analysis (n = 601). Also, 136 infants born to the immunised groups were not included in the analysis because they were born before mothers could receive the first dose of the vaccine. Overall, 1182 infants were included in the analysis. A total of 9823 births were considered from the two studies combined.” Demicheli, Barale and Rivetti 2013, Pgs. 7-8

• 46
  • Table 1, Newell et al. 1966, Pg. 866
  • "The study commenced with the survey and registration of all women between the ages of 13 and 45 years who could be contacted; the formation of a pregnancy register kept up to date by continuous survey; the encouragement of birth and death registration; and the regular visiting of the families of all known or suspected newborn children (at birth, at suspected tetanus illness, at death, and/or at 30 days after birth) to complete birth registration, and to record and distinguish between tetanus and non-tetanus death in the first 30 days of life…In the immunization phase, all registered women were allotted a code number according to their order of ascertainment. These code numbers had previously been divided by means of random sampling numbers into two groups, A and B. Women who were registered, but who refused any immunization, were placed in group C. Any woman who agreed to accept immunization was injected intramuscularly with 1 ml of a preparation bearing the same group letter (i.e., A or B) as her registration card. One vaccine was a 10 LF commercial aluminium-phosphate-adsorbed tetanus toxoid and the other a polyvalent influenza-virus vaccine…Three injections were offered to each woman with a minimum interval of 6 weeks between injections.” Newell et al. 1966, Pg. 865

• 47

  At least 2 doses: (145+372)/772. At least 3 doses: 372/772. Table 1, Newell et al. 1966, Pg. 866

• 48

  "During July and August 1974, the protective effect of a glutaraldehyde-treated cholera toxoid was evaluated in the Matlab field study area of the International Centre for Diarrhoeal Disease Research (ICDDR), Bangladesh (formerly the Cholera Research Laboratory) (8).Children between the ages of 1 and 14 years and non-pregnant women at least 15 years old were vaccinated after their informed consent had been obtained. On a double-blind basis, volunteers received 0.5 ml of cholera toxoid or 0.5ml of adult-dose aluminum phosphate-adsorbed Td toxoids (Wyeth Laboratories, Inc.) by intramuscular injection from a Ped-O-Jetainjector. Attempts were made 42 days later to give all those vaccinated a second injection of the same vaccine, and 74% actually received a second dose. One or two injections of cholera toxoids were given to 46443 persons, one injection of Td toxoid was given to 13220 persons, and two injections of Td toxoids were given to 33 175 persons.” Black, Huber and Curlin 1980, Pgs. 927-928

• 49
  • See Studies of mass immunization campaigns to combat MNT (xlsx) for all the studies identified by Demicheli, Barale and Rivetti 2013 and why we excluded some of the studies
  • “Because of the limited number of eligible experimental studies included in this review, we decided to carry out an extended search in order to retrieve all the available comparative studies on this topic. The same databases were explored in order to identify cohort, case-control studies and other non-randomised study designs. Seven further studies were identified (four surveys comparing the disease incidence before and after the introduction of the immunisation campaign, two case-control studies and a cohort study).” Demicheli, Barale and Rivetti 2013, Pg. 11
  • Table 2, Demicheli, Barale and Rivetti 2013, Pg. 29
  • “All the studies but one confirmed the existence of a significant protective effect of an immunisation course of at least two doses of tetanus toxoid on the incidence of neonatal tetanus.” Demicheli, Barale and Rivetti 2013, Pg. 11

• 50
  • We don’t have much detail on Yusuf et al. 1991, because we couldn’t find the study online.
  • “Neonatal tetanus mortality declined dramatically in Pidie district (Aceh Province) Indonesia between 1984 and 1987. Baseline and follow-up survey results demonstrated an 85% reduction in neonatal tetanus mortality during this period, from 32.1/1000 live births to 4.9/1000 live births. During 1985 a tetanus toxoid mass campaign was conducted in Pidie district which resulted in 84% of women 10-45 years of age receiving two tetanus toxoid injections. Analysis of the results of the two surveys provides very strong evidence of the impact of the tetanus toxoid mass campaign on neonatal tetanus mortality.” Abstract, Yusuf et al. 1991

• 51
  • “Neonatal tetanus mortality declined dramatically in Pidie district (Aceh Province) Indonesia between 1984 and 1987. Baseline and follow-up survey results demonstrated an 85% reduction in neonatal tetanus mortality during this period, from 32.1/1000 live births to 4.9/1000 live births. During 1985 a tetanus toxoid mass campaign was conducted in Pidie district which resulted in 84% of women 10-45 years of age receiving two tetanus toxoid injections. Analysis of the results of the two surveys provides very strong evidence of the impact of the tetanus toxoid mass campaign on neonatal tetanus mortality.” Abstract, Yusuf et al. 1991
  • "Estimates of immunization coverage are generally based on two sources of empirical data: reports of vaccinations performed by service providers (administrative data**) and household surveys containing items on children's vaccination history (coverage surveys)...It is important to distinguish whether data accurately reflect immunization system performance or whether they are compromised and thus present a misleading view of immunization coverage. Officially reported data are compared with independent surveys." Immunization Country Reports, Childinfo
  • "The quality of administrative data on immunisation coverage remains suspect due to problems with measurement,[13] as well as the potential for target-oriented initiatives such as UCI[2,14,15] and performance-based payment systems such as GAVI's ISS[2,7] to encourage health-service providers to over-report coverage." Lim et al. 2008, Pg. 2032. We are not sure if the quality of SIA immunization coverage estimated from administrative sources generally differs from the quality of routine immunization coverage estimated from administrative sources.

• 52

  “In China during 1995–1996 widespread tetanus toxoid (TT) mass vaccination of women of childbearing age in high-risk areas was conducted and neonatal tetanus (NT) surveillance was initiated as part of NT elimination efforts.” Chai et al. 2004, Pg. 551

• 53
  • "The national annual incidence of reported NT was 0.18/1000 LB (n = 3657 case) in China in 1996. After an initial increase to 0.21/1000LB in 1997, incidence decreased between 1997 and 2001 from 0.21/1000 LB (n = 4394 cases) to 0. 16/1000 LB (n = 2814 cases) (Table 1). The sensitivity of the surveillance system increased following its implementation: the number and proportion of counties reporting at least one case increased from 627 (24.2%) in 1996 to 904 (35.1%) in 1997 and 894 (34.9%) in 1998. Following this initial increase, the proportion of counties reporting at least one case decreased from 1999 to 2000, concurrent with the decline in incidence.” Chai et al. 2004, Pg. 553
  • Table 1, Chai et al. 2004, Pg. 552
  • “NT has been included in the National Notifiable Disease Reporting System (NNDRS) since 1996. It is a passive reporting system which receives reports primarily from hospitals, although all health care providers, including midwives, are legally required to report NT cases. Cases are reported hierarchically from the village up through to the township, county, provincial, and then to the national level. Any counties with a case of NT are required to report (Table 1), therefore, reporting counties reflect those counties with at least one case of NT. Because NT tends to occur focally in poor underserved areas, we assumed that cases reported from these counties represented cases occurring nationally. We analysed all NT case reports received at the national level through this system and calculated national and provincial incidence rates per 1000 LB, using estimates of national and provincial LB from the national bureau of statistics.” Chai et al. 2004, Pg. 552

• 54
  • Chai et al. 2004 claims the sensitivity of the surveillance system increased over time, but it does not seem to be possible to distinguish the effect of the SIAs from changes in the sensitivity of the surveillance system based on the data alone: “This paper presents the first 6-year national analysis of NT surveillance data in China (1996–2001) since the disease became reportable. Although reporting sensitivity increased following implementation of the surveillance system, as demonstrated by an initial increase in the number and proportion of counties reporting, the incidence decreased during this same period. This decline suggests that NT elimination strategies were effective, particularly targeted TT vaccination among women whose infants are at high-risk for NT. Because we could not estimate more precisely the interventions and incidence at the county level, we may be underestimating the impact of these interventions.” Chai et al. 2004, Pg. 555
  • Surveillance systems generally seem to miss a lot of cases. We’re not sure if China is an exception. "Despite increased awareness of neonatal tetanus, surveillance has not improved much, and reporting is still below 10% in countries in which this disease continues to be a substantial public-health problem. Systematic community-based surveys of neonatal tetanus, like those undertaken in the 1970s and 1980s, have not been repeated since the initiation of neonatal tetanus elimination programme. Thus, estimates of the burden of neonatal tetanus and progress in its elimination, derive from mathematical models that compute the yearly incidence and mortality for each country using the baseline rate of neonatal tetanus before introduction of tetanus toxoid and promotion of clean deliveries, with adjustment for the estimated proportion of women immunised with tetanus toxoid and deliveries assisted by trained personnel. Modelled estimates of neonatal tetanus burden probably provide reasonable approximations of continuing worldwide burden and the effect of activities to eliminate maternal and neonatal tetanus. However, variations in assumptions and data sources used in the modelling process have led to large differences in estimates, with broad uncertainty intervals that highlight the limitations of the available data. Improved surveillance of maternal and neonatal tetanus, and systematic community-based surveys, are clearly needed to validate assumptions used in the models, and to improve the accuracy of present estimates of burden." Roper, Vandelaer, and Gasse 2007, Pg. 1955

• 55
  • "During 1995 and 1996 supplemental TT vaccination was conducted in 320 of 560 counties identified as high-risk, reaching an estimated 23 million women aged 18–35. The reported coverage with two or more doses of TT (TT2+) reached in these 320 counties was 80%. [3]” Chai et al. 2004, Pg. 552
  • We have not found an English copy of [3] online.

• 56
  • "Reports on validation of MNTE elimination in countries", Maternal and Neonatal Tetanus (MNT) related publications and useful links
  • "Reports on progress towards MNT elimination", Maternal and Neonatal Tetanus (MNT) related publications and useful links

• 57

  "Once a country concludes that it has met the definition of MNTE, a request is made to WHO headquarters (HQ) to conduct a formal assessment, and usually a community-based neonatal mortality survey is performed in the district at highest risk for MNT to validate the elimination status. If the survey shows that the highest-risk district has met the definition of MNTE of less than 1 NT case per 1,000 live births, it is assumed that all districts at lower risk also have achieved elimination, as has the country as a whole." Achieving and Sustaining Maternal and Neonatal Tetanus Elimination: Strategic Plan 2012-2015, Pg. 17

• 58

  "The quality assurance sampling approach allows a population to be classified as acceptable or not acceptable in respect of a particular attribute. The decision to accept or not accept is based on the mathematical probabilities of obtaining fewer than a specified number of units with a particular attribute in a random sample of predetermined size drawn from the population to be classified. The major advantage of using a quality assurance sampling design is that smaller sample sizes are needed than in conventional sample surveys in which point estimates with designated precisions are desired." Stroh and Birmingham 2002, Pg. 29

• 59
  • Figure 5, Roper, Vandelaer, and Gasse 2007, Pg. 1955
  • "If the survey shows that the highest-risk district has met the definition of MNTE of less than 1 NT case per 1,000 live births, it is assumed that all districts at lower risk also have achieved elimination, as has the country as a whole." Achieving and Sustaining Maternal and Neonatal Tetanus Elimination: Strategic Plan 2012-2015, Pg. 17
  • "The risk of death is highest for very young and very old patients. Neonatal tetanus mortality approached 100% in community-based surveys in the 1980s, but is now 10–60% with hospital care." Roper, Vandelaer, and Gasse 2007, Pg. 1950

• 60
  • These reports describe surveys assessing the neonatal mortality rate in a country ("Reports on progress towards MNT elimination", Maternal and Neonatal Tetanus (MNT) related publications and useful links):
    • "In April 2006, the Government of Nigeria, in collaboration with WHO and UNICEF, conducted a community-based survey to assess the incidence of neonatal tetanus (NT) in Kano State." Report on progress towards MNT elimination, Nigeria (Kano State), 2006, Pg. 433
    • "Assessment of neonatal tetanus mortality in Badakshan, Kandahar and Kunduz provinces, Afghanistan, 2004" Report on progress towards MNT elimination, Afghanistan (Badakshan, Kandahar, Kunduz provinces), 2004, Pg. 225
    • "This report summarizes the results from 11 community-based surveys conducted in 12 districts of Indonesia in February 2001. The objectives of the surveys were to assess progress towards MNT elimination in Indonesia; to assess the methods used to classify districts as low, medium and high NT risk; and to assess TT2+ coverage among recently pregnant women and other CBAWs in selected low-, medium- and high-risk districts." Report on progress towards MNT elimination, Indonesia, 2003, Pg. 330
    • "Assessment of incidence of neonatal tetanus in selected districts of the Lao People's Democratic Republic" Report on progress towards MNT elimination, Lao People's Democratic Republic, 2002, Pg. 277
    • "Assessment of neonatal tetanus in Niger" Report on progress towards MNT elimination, Niger, 2002, Pg. 242
    • "A baseline survey for NT mortality was conducted in each selected county." Report on progress towards MNT elimination, China, 1993, Pg. 202
  • We do not include surveys to validate elimination in a country, because the sampling methods employed by these surveys do not yield precise estimates of mortality: "The quality assurance sampling approach allows a population to be classified as acceptable or not acceptable in respect of a particular attribute. The decision to accept or not accept is based on the mathematical probabilities of obtaining fewer than a specified number of units with a particular attribute in a random sample of predetermined size drawn from the population to be classified. The major advantage of using a quality assurance sampling design is that smaller sample sizes are needed than in conventional sample surveys in which point estimates with designated precisions are desired." Stroh and Birmingham 2002, Pg. 29
  • We do not include "Indonesia, 2003", because the level of neonatal tetanus mortality in the districts surveyed was not reported. Report on progress towards MNT elimination, Indonesia, 2003

• 61

  "In April 2006, the Government of Nigeria, in collaboration with WHO and UNICEF, conducted a community-based survey to assess the incidence of neonatal tetanus (NT) in Kano State." Report on progress towards MNT elimination, Nigeria (Kano State), 2006, Pg. 433
  

• 62"Data were collected between 11 April and 16 April 2006, and data entry was done during 17–18 April 2006." Report on progress towards MNT elimination, Nigeria (Kano State), 2006, Pg. 435
  
• 63

  "Mortality from neonatal tetanus (per 1000 live births)", Table 2, Report on progress towards MNT elimination, Nigeria (Kano State), 2006, Pg. 436

• 64

  "The 3 provinces and the districts were purposely selected in a 3-stage process on the basis of accessibility, security and a perceived high risk for NT. Firstly, provinces with good coverage by health services or with poor access during winter months and/or low security were excluded. Secondly, among the remaining provinces, the NT risk status was assessed, based on reported vaccination coverage of pregnant women with 2 or more doses of TT (TT2+), vaccination coverage of infants with 3 doses of diphtheria–tetanus–pertussis (DTP3) and proportion of births assisted by trained attendants. Thus Badahkshan, Kandahar and Kunduz provinces were selected for inclusion in the survey. Table 1 summarizes the indicator values for the country’s 32 provinces." Report on progress towards MNT elimination, Afghanistan (Badakshan, Kandahar, Kunduz provinces), 2004, Pg. 225
  

• 65

  "In May–June 2004, the Government of Afghanistan, in collaboration with WHO and UNICEF, conducted a community-based survey on the burden of NT in 3 of the country’s 32 provinces." Report on progress towards MNT elimination, Afghanistan (Badakshan, Kandahar, Kunduz provinces), 2004, Pg. 225
  

• 66

  "NT death rate (per 1000 LB)", Table 3, Report on progress towards MNT elimination, Afghanistan (Badakshan, Kandahar, Kunduz provinces), 2004, Pg. 229

• 67

  "In November 2001, the Government of the Lao People's Democratic Republic, in collaboration with WHO, UNICEF, and the National Statistical Centre, conducted a community-based survey to assess the incidence of neonatal tetanus (NT) in 9 purposefully selected districts in 3 provinces. The 9 districts were selected on the basis of limited access to health care and probable high risk of NT." Report on progress towards MNT elimination, Lao People's Democratic Republic, 2002, Pgs. 277-278

• 68

  "A total of 7919 households, containing 46,730 residents, were visited between 2 November and 2 December 2001." Report on progress towards MNT elimination, Lao People's Democratic Republic, 2002, Pg. 278

• 69"Neonatal tetanus mortality per 1000 live births", Table 1, Report on progress towards MNT elimination, Lao People's Democratic Republic, 2002, Pg. 279
• 70"Three districts (Birni N'Konni, Keita and Mirriah) were selected proportional to population size (PPS) to represent the 36 districts in which 96% of the population of Niger resides (the 4 northern districts were excluded from the sampling frame for logistic reasons)." Report on progress towards MNT elimination, Niger, 2002, Pg. 242
• 71

  "The Ministry of Health of the Government of Niger, in collaboration with WHO, conducted an assessment of neonatal tetanus (NT) in Niger during a 3-week period in August and September 2001." Report on progress towards MNT elimination, Niger, 2002, Pg. 242

• 72

  "Estimate of neonatal mortality and neonatal tetanus mortality rates," "NTMR definite," Table 1, Report on progress towards MNT elimination, Niger, 2002, Pg. 245
  

• 73

  "The project covers a population of 120 million people in 300 selected counties (10% of all counties). The selection criteria included high infant mortality and birth rates and low per capita income. The number of counties selected for each province ranged from 3 to 20. A baseline survey for NT mortality was conducted in each selected county. Townships were the unit of sampling and typically 5 or 6 townships out of 20 or so in a county were chosen. Within each township selected, in each village the village leader and health workers listed all births, including unregistered ones, that had occurred in the previous year (1989), and separately registered all infants and 1-4-year-old child deaths. The causes of infant deaths recorded included: asphyxia/birth trauma, neonatal tetanus, pneumonia, diarrhoea, prematurity, injury/accident, and others. The provincal figures from the baseline survey are not necessarily indicative of the whole province, and are expected to be higher than the average for the whole province because of selection of the areas at high risk." Report on progress towards MNT elimination, China, 1993, Pg. 202

• 74

  "Infant mortality rate, neonatal mortality and neonatal tetanus mortality rates per 1000 live births; results of the baseline survey in 300 counties (sampling villages), China, 1989", Table 2, Report on progress towards MNT elimination, China, 1993, Pg. 203

• 75

  “Neonatal tetanus mortality rate per 1000 live births”, Table 2, Report on progress towards MNT elimination, China, 1993, Pg. 203

• 76

  World Bank Data, Under-5 mortality

• 77
  • Cluster sampling
    • "The protocol was a slightly modified version of the WHO cluster survey method developed to establish baseline estimates of the incidence of NT." Report on progress towards MNT elimination, Nigeria (Kano State), 2006, Pg. 434
    • "The survey design was adapted from the WHO cluster survey developed to establish baseline estimates for mortality attributable to NT." Report on progress towards MNT elimination, Afghanistan (Badakshan, Kandahar, Kunduz provinces), 2004, Pg. 226
    • "The survey design was a slightly modified version of the WHO cluster survey developed to establish baseline estimates of NT incidence; the total sample size was reduced and the number of clusters was increased." Report on progress towards MNT elimination, Lao People's Democratic Republic, 2002, Pg. 278
    • Report on progress towards MNT elimination, Niger, 2002 does not mention a WHO protocol. This survey randomly sampled districts from most of the country rather than targeting high-risk districts. "The Ministry of Health of the Government of Niger, in collaboration with WHO, conducted an assessment of neonatal tetanus (NT) in Niger during a 3-week period in August and September 2001. A community-based survey was conducted to assess the burden of NT and tetanus toxoid (TT) coverage among mothers. In addition, records were reviewed at the national statistical centre (SNIS, “Service National d’Information Sanitaire”) and in hospitals and health offices at district and regional levels to evaluate surveillance for NT. The information obtained during the assessment was useful in the development of the Ministry of Health’s 5-year plan of action to eliminate NT. Three districts (Birni N’Konni, Keita and Mirriah) were selected proportional to population size (PPS) to represent the 36 districts in which 96% of the population of Niger resides (the 4 northern districts were excluded from the sampling frame for logistic reasons). Based on the crude birth rate (CBR) and the average household size, a second- stage PPS sample of 80 cluster locales was selected. Each cluster was to consist of 25 live births (LB), a size that would allow each cluster to be completed in one day’s work by an interview team. Data collected during the survey were recorded on 4 forms. Form 1 recorded information on household size and identified households in which LB occurred during the period 1 September 2000 to 31 August 2001. Information on LB that occurred during this period, such as date of birth, sex, survival status and age at death, were recorded on Form 2. Form 3 was used to record information on a sub-sample of the first 5 mothers of LB in each cluster. Mothers were asked their age, whether they had a vaccination card, the number of doses of TT received, the place where their baby was delivered and whether a health care worker (HCW) had assisted at the delivery. A question was also asked about where the mother would seek care if her neonate became ill. Form 4 recorded detailed information on neonatal deaths (ND), including signs of illness preceding death, to allow diagnosis of NT by verbal autopsy. Deaths possibly due to NT were classified on the basis of normal suckling during the first 2 days of life, cessation of suckling, any clinical sign of NT, and death after the third day of life. A definite case of NT included the above, plus a history of 1 or more specific clinical signs of NT and at least 2 risk factors for NT (birth at home, inadequate TT immunization, no antenatal care visits, etc.). " Pg. 242
    • "A project initiated in 1990 by the Maternal and Child Health Department of the Ministry of Public Health and supported by UNICEF and the United Nations Fund for Population Activities (UNFPA) seeks to strength the quality of health care at the grassroots level and to reduce infant and child mortality. The project covers a population of 120 million people in 300 selected counties (10% of all counties). The selection criteria included high infant mortality and birth rates and low per capita income. The number of counties selected for each province ranged from 3 to 20. A baseline survey for NT mortality was conducted in each selected county. Townships were the unit of sampling and typically 5 or 6 townships out of 20 or so in a county were chosen. Within each township selected, in each village the village leader and health workers listed all births, including unregistered ones, that had occurred in the previous year (1989), and separately registered all infants and 1-4-year-old child deaths. The causes of infant deaths recorded included: asphyxia/birth trauma, neonatal tetanus, pneumonia, diarrhoea, prematurity, injury/accident, and others. The provincal figures from the baseline survey are not necessarily indicative of the whole province, and are expected to be higher than the average for the whole province because of selection of the areas at high risk." Report on progress towards MNT elimination, China, 1993, Pg. 202
  • "Three districts (Birni N’Konni, Keita and Mirriah) were selected proportional to population size (PPS) to represent the 36 districts in which 96% of the population of Niger resides (the 4 northern districts were excluded from the sampling frame for logistic reasons)." Report on progress towards MNT elimination, Niger, 2002, Pg. 242
  • "Kano is one of the largest of Nigeria’s 36 states, with a population of 8.9 million. Administratively, it consists of 44 local government areas and 483 wards...A total of 108 wards were selected, proportional to their population size, from the list of 483 wards. Within each selected ward, 1 settlement was selected as the cluster site; this was done by choosing the settlement that fell at the midpoint in an alphabetical listing of all settlements within that ward. Of the 108 clusters, 34 were located in the 8 local government areas that form the Kano metropolitan area." Report on progress towards MNT elimination, Nigeria (Kano State), 2006, Pgs. 433-434
  • "The 9 districts were selected on the basis of limited access to health care and probable high risk of NT." Report on progress towards MNT elimination, Lao People's Democratic Republic, 2002, Pg. 278
  • "The 3 provinces and the districts were purposely selected in a 3-stage process on the basis of accessibility, security and a perceived high risk for NT." Report on progress towards MNT elimination, Afghanistan (Badakshan, Kandahar, Kunduz provinces), 2004, Pg. 225
  • "The selection criteria included high infant mortality and birth rates and low per capita income. The number of counties selected for each province ranged from 3 to 20. A baseline survey for NT mortality in each selected county. Townships were the unit of sampling and typically 5 or 6 townships out of 20 or so in a county were chosen...The provincal figures from the baseline survey are not necessarily indicative of the whole province, and are expected to be higher than average for the whole province because of selection of the areas at high risk." Report on progress towards MNT elimination, China, 1993, Pg. 202

• 78

  See the “Table: Surveys assessing neonatal tetanus mortality rates” above

• 79
  • Table 1, Million deaths study 2010, Pg. 14
  • “The Registrar General of India conducted a survey of all deaths occurring in 2001-03 in 1.1 M nationally representative homes. About 800 field staff interviewed households and completed standard questions and a half-page narrative about the events that preceded the death. Each field report was sent to two of 130 trained physicians, who independently assigned an ICD-10 code to each death. Discrepancies were resolved via anonymous reconciliation and, if necessary, adjudication. Cause-specific mortality rates for 2005 were calculated nationally and for the six regions by combining the observed proportions for each cause among 10 892 deaths in neonates and 12 260 deaths at ages 1-59 months with United Nations population and death totals.” Million deaths study 2010, Pg. 1

• 80
  • “Since its inception in 1990 the neonatal tetanus elimination initiative has stressed focus on communities traditionally missed by routine immunisation and maternal and child health services, because of geographical and sociocultural barriers to health-service access. These high-risk communities typically have a disproportionately high neonatal tetanus burden, and high infant and maternal mortality in general. As the programme matured, guidelines were developed for systematically identifying high-risk districts or areas with the algorithm in figure 5." Roper, Vandelaer, and Gasse 2007, Pg. 1954
  • Figure 5, Roper, Vandelaer, and Gasse 2007, Pg. 1955
  • "Despite increased awareness of neonatal tetanus, surveillance has not improved much, and reporting is still below 10% in countries in which this disease continues to be a substantial public-health problem. Systematic community-based surveys of neonatal tetanus, like those undertaken in the 1970s and 1980s, have not been repeated since the initiation of neonatal tetanus elimination programme. Thus, estimates of the burden of neonatal tetanus and progress in its elimination, derive from mathematical models that compute the yearly incidence and mortality for each country using the baseline rate of neonatal tetanus before introduction of tetanus toxoid and promotion of clean deliveries, with adjustment for the estimated proportion of women immunised with tetanus toxoid and deliveries assisted by trained personnel. Modelled estimates of neonatal tetanus burden probably provide reasonable approximations of continuing worldwide burden and the effect of activities to eliminate maternal and neonatal tetanus. However, variations in assumptions and data sources used in the modelling process have led to large differences in estimates, with broad uncertainty intervals that highlight the limitations of the available data. Improved surveillance of maternal and neonatal tetanus, and systematic community-based surveys, are clearly needed to validate assumptions used in the models, and to improve the accuracy of present estimates of burden." Roper, Vandelaer, and Gasse 2007, Pg. 1955

• 81
  • “Births assisted by health workers, within or outside health facility.” Table 3, Report on validation of MNTE elimination in countries, Nepal, 2006, Pg. 125
  • The coverage rates under “Vaccination coverage among CBAW based on card and history” sum to more than 100%, so we assume the coverage rates refer to the proportion of women of childbearing age with at least a certain number of doses (rather than exactly a certain number of doses). “Tetanus toxoid (TT) coverage of mothers (n = 438) and women of childbearing age (CBAW) (n = 438) in Banke, Solukhumbu and Syangia districts, Nepal, 2005.” Table 4, Report on validation of MNTE elimination in countries, Nepal, 2006, Pg. 126
  • “A finding of more than 2 NT deaths in this survey would have supported the conclusion that NT had not been eliminated in the surveyed districts. However, as only 1 death attributable to NT was found, the results suggest that NT had been eliminated in the 3 districts at the time of the survey. As the 3 districts were purposefully selected because they were judged to be at highest risk of NT, it is likely that NT has also been eliminated in the rest of the country, which was considered to be at lower risk of NT at the time of the survey.
    The neonatal death rate estimated from the survey is 7.4 per 1000 LBs, a rate substantially below the WHO estimate of 40 per 1000 for the year 2000.3 The survey estimate of the rate of stillbirths and third-trimester abortions was 26.7 per 1000 births (95% CI 19–34 per 1000); that rate is comparable to the WHO estimate of 23 stillbirths per 1000 total births during 2000.3 It is possible that some early neonatal deaths were reported by the respondents as stillbirths or third-trimester abortions, which may partially explain the low rate of neonatal deaths found in the survey. NT mortality surveys have often produced estimates of neonatal death rates lower than expected. The reasons may vary from country to country, and may include cultural, survey- related or other causes. It is, however, also possible that the trend of declining neonatal mortality in Nepal is continuing, resulting in neonatal mortality rates that are currently below the WHO estimate for 2000.” Report on validation of MNTE elimination in countries, Nepal, 2006, Pgs. 125-126
  • Though the survey didn’t report the coverage rate of at least 2 doses of TT vaccine among pregnant women, it reported a coverage rate among mothers of 74.8% (Table 4, Report on validation of MNTE elimination in countries, Nepal, 2006, Pg. 126). The report describing the survey also cites administrative data that showed TT2 coverage of pregnant women at 20% in Syangja, 7% in Solukhumbu, and 62% in Banke (Table 2, Report on validation of MNTE elimination in countries, Nepal, 2006, Pg. 122). This administrative data may have problems though: “The TT2 coverage estimates from the survey, based on card and history, are higher than the estimates of coverage from administrative reports. Health workers often ‘re-start’ the TT schedule if the women cannot show a vaccination card, and therefore underreport coverage of TT2+. The survey confirmed the low card retention rates, and a higher than reported TT coverage. In addition, HMIS reports tend not to include coverage obtained through SIAs. Misreporting by respondents on their vaccination history is also possible, but other studies have shown that such misreporting tends to be minimal. The higher estimates, however, support the finding that NT had been eliminated in the districts surveyed.” Report on validation of MNTE elimination in countries, Nepal, 2006, Pgs. 126-127

• 82
  • The median coverage rate of at least 2 doses of TT vaccine among mothers in surveys described in MNTE Initiative reports that found a neonatal tetanus mortality rate greater than or equal to 1 death per 1,000 live births (both assessment surveys and failed validation surveys) is 49% (interquartile range: 31% to 51%).
    • See ≥1 tab in MNT coverage data (xlsx)
    • We assume that the coverage rates refer to the proportion of mothers with at least a certain number of doses rather than a certain number of doses exactly, because in a few surveys that the coverage rates sum to more than 100% (e.g. “Vaccination coverage among mothers based on card and history”, Report on validation of MNTE elimination in countries, Nepal, 2006, Pg. 126; “Mothers with immunization verified by card and history”, Report on validation of MNTE elimination in countries, Bangladesh, 2008, Pg. 306; Table 6, Report on validation of MNTE elimination in countries, Burundi and the Comoros, 2009, Pg. 297), but we haven’t checked all the surveys and we haven’t seen the coverage rates unambiguously defined.
  • The median coverage rate of at least 2 doses of TT vaccine among mothers in surveys described in MNTE Initiative reports that found a neonatal tetanus mortality rate less than 1 death per 1,000 live births (successful validation surveys) is 78% (interquartile range: 72% to 83%).
    • See <1 tab in MNT coverage data (xlsx)
    • We assume that the coverage rates refer to the proportion of mothers with at least a certain number of doses rather than a certain number of doses exactly, because in a few surveys that the coverage rates sum to more than 100% (e.g. “Vaccination coverage among mothers based on card and history”, Report on validation of MNTE elimination in countries, Nepal, 2006, Pg. 126; “Mothers with immunization verified by card and history”, Report on validation of MNTE elimination in countries, Bangladesh, 2008, Pg. 306; Table 6, Report on validation of MNTE elimination in countries, Burundi and the Comoros, 2009, Pg. 297), but we haven’t checked all the surveys and we haven’t seen the coverage rates unambiguously defined.

• 83

  SIA immunization coverage (xlsx)

• 84

  SIA immunization coverage (xlsx)

• 85
  • "In order to eliminate maternal and neonatal tetanus, between 2002 and 2006, SIAs for TT were implemented in 52 districts considered to be at high risk. Of the 2 330 000 women of childbearing age targeted, 81% had received 2 doses of TT (TT2) and 72%, 3 doses (TT3)." Report on progress towards MNT elimination, Mali, 2007, Pg. 310
  • "Gao had an infant population of 7417, urban dwellers constituted 27% of the total population, and TT coverage in SIAs was 64% (TT2) and 50% (TT3)." Report on progress towards MNT elimination, Mali, 2007, Pg. 310
  • "Supplementary immunization activities delivering >2 doses of tetanus toxoid vaccine in high-risk townships, Myanmar, 1999–2009." Table 1, Report on validation of MNTE elimination in countries, Myanmar, 2010, Pg. 430
  • "TT coverage of women of reproductive age, by round during supplemental immunization activities, Comoros, 2002, 2003 and 2009," Tables 2, and "Results of supplemental tt immunization activities, Burundi, 2007 and 2009," Table 3, Report on validation of MNTE elimination in countries, Burundi and the Comoros, 2009, Pg. 291
  • "Supplementary immunization activities (SIAs) to deliver tetanus–diphtheria (Td) vaccine in 60 provinces, Turkey, 2006–2007," Table 1, Report on validation of MNTE elimination in countries, Turkey, 2009, Pg. 142
  • "Supplementary immunization activities (SIAs) to deliver tetanus toxoid vaccine (TT) in 10 high risk districts, the Congo, 2006," Table 1, Report on validation of MNTE elimination in countries, Congo (the), 2009, Pg. 471
  • "Supplementary immunization activities (SIAs) implemented to deliver tetanus toxoid (TT), Bangladesh, 1999–2006,"
    Table 1, Report on validation of MNTE elimination in countries, Bangladesh, 2008, Pg. 302
  • "Summary of tetanus toxoid (TT) supplementary immunization activities (SIAs) and reported neonatal tetanus (NT) cases, Nepal, 2000–2005," Table 1, Report on validation of MNTE elimination in countries, Nepal, 2006, Pg. 121
  • "Summary of UNICEF-supported supplementary immunization activities (SIAs) to provide tetanus toxoid (TT), Viet Nam, 2002– 2005," Table 1, Report on validation of MNTE elimination in countries, Viet Nam, 2006, Pg. 267
  • "Indicator data for districts identified as being at high risk for neonatal tetanus, Rajastan, 2000." Table 1, Report on progress towards MNT elimination, India (Rajastan), 2003, Pg. 27
  • "The coverage of TT3 attained during SIAs was 67%." , Report on validation of MNTE elimination in countries, Egypt, 2007, Pg. 238

• 86
  • "Coverage estimates based on administrative data", Table 1, Report on progress towards MNT elimination, India (Rajastan), 2003, Pg. 27
  • "Estimates of immunization coverage are generally based on two sources of empirical data: reports of vaccinations performed by service providers (administrative data**) and household surveys containing items on children's vaccination history (coverage surveys)...It is important to distinguish whether data accurately reflect immunization system performance or whether they are compromised and thus present a misleading view of immunization coverage. Officially reported data are compared with independent surveys." Immunization Country Reports, Childinfo
  • "The quality of administrative data on immunisation coverage remains suspect due to problems with measurement,[13] as well as the potential for target-oriented initiatives such as UCI[2,14,15] and performance-based payment systems such as GAVI's ISS[2,7] to encourage health-service providers to over-report coverage." Lim et al. 2008, Pg. 2032. We are not sure if the quality of SIA immunization coverage estimated from administrative sources generally differs from the quality of routine immunization coverage estimated from administrative sources.
  • Some of the coverage rates exceed 100% (See SIA immunization coverage (xlsx)). Report on progress towards MNT elimination, India (Rajastan), 2003 states that “Administrative reports of SIA coverage were calculated using the number of recorded doses administered to target women per round as the numerator and the target population of women as the denominator; the latter was estimated using projections from the 1991 census.” Pgs. 29-30. An out-of-date denominator could lead to an inaccurate estimate of coverage as well as an inflated number of doses delivered.

• 87

  "Once a country concludes that it has met the definition of MNTE, a request is made to WHO headquarters (HQ) to conduct a formal assessment, and usually a community-based neonatal mortality survey is performed in the district at highest risk for MNT to validate the elimination status. If the survey shows that the highest-risk district has met the definition of MNTE of less than 1 NT case per 1,000 live births, it is assumed that all districts at lower risk also have achieved elimination, as has the country as a whole." Achieving and Sustaining Maternal and Neonatal Tetanus Elimination: Strategic Plan 2012-2015, Pg. 17

• 88
  • See “Summary” tab in MNT coverage data (xlsx)
  • We assume that the coverage rates refer to the proportion of mothers with at least a certain number of doses rather than a certain number of doses exactly, because in a few surveys that the coverage rates sum to more than 100% (e.g. “Vaccination coverage among mothers based on card and history”, Report on validation of MNTE elimination in countries, Nepal, 2006, Pg. 126; “Mothers with immunization verified by card and history”, Report on validation of MNTE elimination in countries, Bangladesh, 2008, Pg. 306; Table 6, Report on validation of MNTE elimination in countries, Burundi and the Comoros, 2009, Pg. 297), but we haven’t checked all the surveys and we haven’t seen the coverage rates unambiguously defined.
  • Coverage rates are based on immunization cards and self-report (see MNT coverage data). We have not vetted the methodology of these surveys.

• 89
  • Assuming that 38% (75% - 37%) receive exactly 2 doses and 37% receive exactly 3 doses during the SIA and that 2 doses are about half as effective as 3 doses, then 38%/2 + 37% = 56%.
  • We assume 2 doses are about half as effective as 3 doses based on Newell et al. 1966 suggesting 2-3 doses (mostly 3) reduce neonatal tetanus mortality by 98% over 54 months and Black, Huber and Curlin 1980 suggesting 2 doses reduce 4-14 day mortality by 68% over 32 months (we’d guess the effectiveness would go down over a longer time period). See above.

• 90
  • "During previous assessments a procedure was developed for identifying districts at highest risk for NT. It is based on three steps: 1) consolidating core and surrogate indicators on every district in a spreadsheet (Annex 1); 2) developing a short list of districts potentially at highest risk by ranking the districts according to the values of the indicators; 3) selecting the districts to be surveyed from the short list on the basis of public health judgement and logistical and administrative considerations." Stroh and Birmingham 2002, Pg. 8. As an example of how the protocol was implemented: "Only districts where more than 65% of the population lived in rural areas and TT3 coverage in SIAs was ≤50% were selected. Four districts met these criteria: Abeldara, Tin Essako, Gao and Nioro. The first 2 were ruled out because their population was too small. Gao was selected over Nioro because of its greater rural population and lower immunization coverage (especially TT2) in SIAs...The district where the neonatal tetanus rate was estimated to be among the highest in the country was thus deliberately chosen to prove that, if neonatal tetanus had been eliminated there, the disease had probably also been eliminated in districts with better indicators." Report on progress towards MNT elimination, Mali, 2007, Pg. 310
  • See the section on Do SIAs target populations with high neonatal tetanus mortality rates? for the discussion of the areas in which assessment surveys took place

• 91

  "Tetanus toxoid is stable and can withstand exposure to temperatures of around 20 °C for months and storage at 37 °C for a few weeks without significant loss of potency. However, the vaccine is destroyed in 2 hours at 56 °C. Tetanus toxoid-containing vaccines should be stored at +4 (2-8) °C; vaccines that have been frozen should not be used." WHO 2006, Pg. 202

• 92

  "While TT immunization and delivery in health facilities are usually seen as protective factors in preventing NT, it is interesting to note that in both surveys an important proportion of mothers of NT cases claimed to have received ≥ 2 TT doses: 16.7% in 1990 and 18.8% in 2006. This apparent contradiction may be linked to poor vaccination techniques (for example, the use of a vaccine that is no longer effective as a result of extended cold-chain interruptions) or errors in recall during the survey." Report on progress towards MNT elimination, Nigeria (Kano State), 2006, Pg. 439

• 93
  • Roper, Vandelaer, and Gasse 2007 (authors from the WHO and UNICEF: See sidebar on Pg. 1947) claims the longer the interval between doses, the greater the immune response: "The longer the interval between doses, the greater the antibody response to the second dose; an interval of at least 6 weeks is recommended when feasible. Maternal tetanus antibody transfer peaks at 60 or more days after the second dose, which should be given several weeks before delivery to ensure protective antibody concentrations in newborn babies.” Roper, Vandelaer, and Gasse 2007, Pg. 1951
  • The WHO also warns against delaying immunization for too long: "Whereas protection is incomplete after the first vaccine dose, protective concentrations of antitoxin are achieved in the majority of vaccines after completion of 2 doses; a third dose induces immunity in almost 100% of those immunized. The interval between the tetanus toxoid-containing doses should be at least 4 weeks. Longer intervals may increase the magnitude and duration of the immune response, but should not be a reason to delay immunization.” WHO 2006, Pg. 203
  • The WHO recommends that SIAs consist of 3 rounds of immunization with a minimum of four weeks between the first and second rounds and a minimum of six months between the second and third rounds: “2. Implementation of three rounds of TT-SIAs in all of the high-risk districts to immunize all women of reproductive age, irrespective of their previous vaccination status. 3. Minimum of four weeks interval between the first and second rounds and a minimum of six months between the second and third rounds.” Achieving and Sustaining Maternal and Neonatal Tetanus Elimination: Strategic Plan 2012-2015, Pg. 11
  • Note that Newell et al. 1966 (the only RCT we've seen that used 3 doses) appears to have used a shorter period between the 2nd and 3rd dose in some cases, yet no babies born to women over the following 54 months died of tetanus in the group of women that received doses on this schedule.
    • “The study group was given 1-3 injections of I ml of an aluminium-phosphate-adsorbed tetanus toxoid more than 6 weeks apart, and the control group a similar number of injections of an influenza-virus vaccine.” Newell et al. 1966, Pg. 863
    • “Interval from injection to birth (months)” and “Tetanus mortality”. Tables 5-6, Newell et al. 1966, Pg. 869
    • See Are the babies born to vaccinated women protected?

• 94
  • "Risk was not reduced by a maternal history of two doses of tetanus toxoid (TT2), although estimated efficacy of TT2 was 45% (95% confidence interval = 16% to 64%). Subsequent to the survey, a reference laboratory reported no potency in three consecutive lots of tetanus vaccine from the production laboratory in Bangladesh." Hlady et al. 1992, Pg. 1365
  • "Subsequent to our findings, a WHO reference laboratory report no potency in samples from three consecutive lots of tetanus toxoid produced in Bangladesh in July 1990 (J. Milstien, WHO, personal communication, October 17, 1990), and a serosurvey of 100 women vaccinated in September and October 1990 with two doses of tetanus toxoid vaccine found that only 31% had developed protective levels of antibody (N. Abiprojo, UNICE, personal communication, July 29, 1991).
    Since 1983, routine potency testing of tetanus toxoid in Bangladesh has been performed in the same facility by the same staff that produced all the tetanus toxoid vaccine for the national immunization program. There is no national control authority in Bangladesh to certify vaccine safety or potency. Currently, tetanus toxoid production in Bangladesh has been suspended, and the immunization program has continued with the use of imported vaccine." Hlady et al. 1992, Pg. 1368

• 95

  “Neonatal tetanus (NT) is a major cause of mortality in developing countries, with over 400000 deaths estimated to occur annually. WHO has adopted the goal of eliminating NT worldwide, and a major strategy for its prevention is the administration of at least two properly spaced doses of tetanus toxoid (TT) to women of childbearing age in high-risk areas to protect passively their newborns at birth. In certain countries the locally produced TT vaccine has been shown to be subpotent, while other countries have reported NT among infants born to vaccinated women. An extensive review of production and quality control procedures was carried out between 1993 and 1995 in 8 of 22 TT-producing countries that also report NT cases, with a more superficial assessment being carried out in the remaining 14 countries. Only 4 of the 22 countries have a functioning national control authority to monitor TT production and vaccine quality. A total of 80 TT lots from 21 manufacturers in 14 of the 22 NT-reporting countries were tested for potency. Of these, 15 lots from eight manufacturers in seven countries had potency values below WHO requirements. TT potency can also be compromised by improper vaccine handling. To eliminate neonatal tetanus worldwide requires assurance that all doses of TT meet WHO production and quality requirements and that the field effectiveness of TT is monitored through systematic NT case investigations and assessment of coverage.” Dietz et al. 1996, Pg. 619

• 96

  "Concerns have been raised that maternal malaria could affect neonatal protection by reduction of maternal response to immunisation or placental antibody transfer. In some studies, malaria infection has been seen to decrease the antibody response to tetanus toxoid in children, although malaria chemoprophylaxis seems to preserve the response. A study comparing pregnant women with and without malaria parasitaemia noted no difference in the antibody response to tetanus toxoid; however, all participants received chloroquine prophylaxis. Studies investigating the effects of placental malaria on transplacental tetanus antibody transfer have had conflicting results: one showed reduced tetanus antibody concentrations in newborn babies of mothers with severe placental malaria, yet two others detected no such effect.

  Most studies of the immune response to tetanus toxoid in HIV-infected patients have been done in children and non-pregnant adults. Infants and adults infected with HIV generally do mount a protective response to tetanus toxoid, but their antibody levels tend to be lower than those of uninfected controls, especially in those whose CD4 lymphocyte counts are less than 300 cells per μL. With disease progression, immune response and serum tetanus antibody concentrations decrease. This blunted response suggests that HIV-infected individuals might need more frequent booster doses. Placental transfer of tetanus IgG was significantly lower in HIV-infected Brazilian mothers than in uninfected controls, a finding that was not replicated in a subsequent study in Malawi. In both studies, all neonates had protective tetanus antibody concentrations." Roper, Vandelaer, and Gasse 2007, Pgs. 1951-1952

• 97
  • "This study aimed to estimate the incremental cost-effectiveness of supplementary immunization activities to prevent neonatal tetanus in the Loralai district of Pakistan. The supplemental immunization activities were carried out in two phases during 2001–03." Griffiths et al. 2004, Pg. 643
  • "Loralai is one of 26 districts in the province of Balochistan, the least developed province in Pakistan (15)…access to health services, including vaccination with tetanus toxoid, is limited in Loralai." Griffiths et al. 2004, Pg. 644.

• 98
  • "The costs of the supplementary immunization activities in US$ are illustrated in Table 3. The total costs (phases 1 and 2) amounted to US$ 26 108. UNICEF procured the tetanus toxoid vaccine at a cost of US$ 0.04 per dose, including freight charges; the vaccine accounted for 11% of total costs. Syringes were considerably more expensive than the vaccine (accounting for 23% of total costs), whereas per diems amounted to 16% of total costs and salaries to 14%. Hence, the Government of Pakistan (through salary costs) funded approximately 14% of the total. Since 65 877 doses of vaccine were delivered during the supplementary activities, the cost per dose amounted to only US$ 0.40. The cost per woman vaccinated with three doses was US$ 1.19." Griffiths et al. 2004, Pgs. 646-647.
  • "The supplemental immunization activities were carried out in two phases during 2001–03." Griffiths et al. 2004, Pg. 643
  • CPI Inflation Calculator, converting 2002 to 2013 dollars.

• 99
  • "This paper compares the cost per completed maternal tetanus immunization and an estimate of the cost per death averted in the routine EPI program with similar results from an experimental mass campaign in Aceh Province, Indonesia." Berman et al. 1991, Pg. 185
  • "Immunizations were given in two rounds beginning in August 1985…The campaign was completed by the end of November 1985." Berman et al. 1991, Pg. 186

• 100
  • Table 1, Berman et al. 1991, Pg. 188
  • CPI Inflation Calculator, converting 1985 to 2013 dollars.

• 101

  We make the following assumptions in our cost-effectiveness analysis of SIAs:

  • We based the program costs on the cost per woman targeted, not the cost per woman immunized. As a result, the program doesn't save money if it fails to achieve a high coverage rate in a campaign.
  • The program reaches the same women in each round, so every woman reached receives 3 doses. This assumption simplifies the calculation. It also models the goal of the “high-risk” approach to SIAs ("High-Risk Area Approach to MNTE 1. Identification of high-risk districts/areas in a country, based on a thorough data review. 2. Implementation of three rounds of TT-SIAs in all of the high-risk districts to immunize all women of reproductive age, irrespective of their previous vaccination status. 3. Minimum of four weeks interval between the first and second rounds and a minimum of six months between the second and third rounds. 4. Support local initiatives to promote clean delivery and clean cord care practices." Achieving and Sustaining Maternal and Neonatal Tetanus Elimination: Strategic Plan 2012-2015, Pg. 11). Nevertheless, one could also plausibly assume that each woman has some probability of receiving a dose in each round, so different women might receive a different number of doses from the campaign. At relatively high coverage rates, the different assumptions don’t substantially change the cost-effectiveness: under the assumption that each woman has a 70% probability of receiving a dose in each round, the probability of receiving 0 doses from the campaign is (1-70%)*(1-70%)*(1-70%) = 2.7%, the probability of receiving 1 dose is 3*(70%)*(1-70%)*(1-70%) = 18.9%, the probability of receiving 2 doses is 3*(70%)*(70%)*(1-70%) = 44.1% and the probability of receiving 3 doses is (70%)*(70%)*(70%) = 34.3%. Assuming that 2 doses provide roughly half the benefit of 3 doses based on Newell et al. 1966 suggesting 2-3 doses (mostly 3) reduce neonatal tetanus mortality by 98% over 54 months and Black, Huber and Curlin 1980 suggesting 2 doses reduce 4-14 day mortality by 68% over 32 months (we’d guess the effectiveness would go down over a longer time period) (see Are the babies born to vaccinated women protected?), that translates into coverage of all women with 3 doses of about 56% (compared to 70%). At lower coverage rates, the assumption used becomes more important. In this case, the cost-effectiveness may also depend on the number of doses previously received by women in the target population. If different women get immunized in each round, then the cost-effectiveness depends on the number of doses previously received by women in the target population. For instance, if some women had recently received one dose, then even if they only get one additional dose during the campaign, that dose could provide immunity (see Are the babies born to vaccinated women protected?). But if some women have had no previous doses, then the campaign might have much less of an effect on them if only one round reaches them (see Are the babies born to vaccinated women protected?).
  • The neonatal tetanus mortality rate reflects the average risk of the women immunized by the campaign. In practice, this risk may have to be inferred based on available data, such as pre-existing immunization coverage rates, clean delivery rates, and skilled birth attendance rates, and on assumptions about how the risk in the area may have changed in the time since data collection and how the risk would continue to change in the absence of an immunization campaign. It may also depend on whether the campaigns tend to immunize populations similar to the populations represented in the available data (e.g. campaigns may consistently immunize easily reached women whereas surveys on coverage and clean birth rates may sample a population more representative of the area targeted).
  • The CEA does not include the impact of SIAs on maternal mortality. The Global Burden of Disease Study estimates that around 3,000 women between the ages of 15 and 60 and around 47,000 neonates died of tetanus in the developing world in 2010 (see Tetanus deaths by age, female, GBD Compare and Tetanus deaths by age, GBD Compare)

• 102
  • "Tetanus toxoid is one of the most extensively used antigens in vaccinations. The acceptance of tetanus immunization demonstrates its known effectiveness and excellent safety profile. Mild local reactions are relatively common following immunization with tetanus toxoid, but severe adverse events are extremely rare." Borrow, Balmer and Roper 2006, Pg. 29
  • “Tetanus toxoid has been regarded as safe and useful since Descombey first reported its production in 1924 (Descombey 1924). Tetanus toxoid consists of a formaldehyde-treated toxin, which after a primary series of properly spaced doses, stimulates the production of antitoxin which protects against tetanus toxin. Local adverse events (erythema, induration, pain at the injection site) are common but usually self-limited. Occasionally exaggerated local reactions are also reported (extensive painful swelling), most often in adults. Severe systemic reactions such as generalised urticaria, anaphylaxis or neurological complication are rare.” Demicheli, Barale and Rivetti 2013, Pg. 3
  • “Neither of the included studies reported adverse effects. Among the excluded studies we identified two studies evaluating the safety of tetanus toxoid. One was carried out in order to evaluate of the safety of different types of vaccine’s adjuvants (MacLennan 1965) and the second was a case-control study assessing the association between vaccination and congenital anomalies (Silveira 1995). In one further study (Salama 2009), pregnant women experienced pain at injection site more frequently after combined tetanus diphtheria administration than after tetanus toxoid alone. Their characteristics are described in the additional Table 1.” Demicheli, Barale and Rivetti 2013, Pg. 10

• 103

  "The reuse of disposable syringes and needles is widespread and contributes significantly to the transmission of hepatitis B and C and HIV. The autodisabled syringe prevents reuse, and disposal in safety boxes reduces the risk to health staff and the general public from contaminated syringes and needles." Jamison et al. eds. 2006, Pg 1335

• 104

  "Data from WHO show that although routine immunization programs (in Bangladesh, Cote d’Ivoire and Morocco) continued at normal levels of funding after the introduction of NIDs, the logistical requirements associated with NIDs put extraordinary demands on personnel and funds allocated to EPI (WHO 2002d). Empirical evidence of direct effects of intensive polio eradication campaigns on routine immunization is weak but there are reports showing that routine immunization services often stop during NIDs as resources and health personnel are transferred to well publicized and intensive campaigns (Macedo and Melgaard 2000). Studies in Tanzania, Nepal and six countries in the Americas have shown that NIDs divert public focus, donors’ attention and resources from routine immunization programs. (Pan America Health Organization 1995; Mogedal and Stenson 1999). NID campaigns distorted established national policies and priorities for health development set by policymakers in the countries studied. The problem is similar to the fungibility of funds issues faced by aid agencies. Health ministries may perceive NIDs as substitutes for routine immunization and decrease allocations of resources accordingly. Because ‘frontline” service providers are rarely appreciated by communities they serve, in some countries like Nigeria, NIDs are attractive to health personnel because of the perceived social exposure that comes with the NIDs campaigns." Msuya 2004, Pgs. 11-12

• 105

  "In all developing countries, public financing of health in constant US$ from domestic sources increased by nearly 100% (IMF 120%; WHO 88%) from 1995 to 2006. Overall, this increase was the product of rising GDP, slight decreases in the share of GDP spent by government, and increases in the share of government spending on health. At the country level, while shares of government expenditures to health increased in many regions, they decreased in many sub-Saharan African countries. The statistical analysis showed that DAH to government had a negative and significant effect on domestic government spending on health such that for every US$1 of DAH to government, government health expenditures from domestic resources were reduced by $0.43 (p=0) to $1.14 (p=0). However, DAH to the non-governmental sector had a positive and significant effect on domestic government health spending. Both results were robust to multiple specifications and subset analyses. Other factors, such as debt relief, had no detectable effect on domestic government health spending." Lu et al. 2010, Pg. 1375

• 106

  "The past few years have seen increased advocacy for vertical programs for a number of reasons. For donors and political establishments, vertical programs are attractive because they show quick results and they are easier to manage than horizontal programs. However, most policy makers in developing countries see vertical programs diverting human and financial resources from already resource-constrained health systems (Schreuder and Kostermans, 2001). One example can be found in vertical programs undertaken as part of “National Immunization Days NIDs” which are supplemental to horizontal programs that include routine vaccination services. A study of immunization services in Southern Africa which focused on poliomyelitis eradication programs, showed that polio programs divert resources and attention from comprehensive Primary Health Care (PHC) Services . On the other hand, the study showed that NIDs are cost-effective and show good results – with respect to immunization coverage. Some argue that NIDs do promote PHC services but evidence supporting this argument is very weak. Questions remain on the long-term sustainability of vertical programs in terms of outcomes and resources. " Msuya 2004, Pg. 3

• 107

  Other immunization services that require traveling to a clinic, for instance, may result in lower coverage than house-to-house immunization campaigns:

  "Four factors were independently associated with being more likely to receive complete immunization with tetanus toxoid (compared with receiving incomplete or no immunization with tetanus toxoid): registration of the pregnant woman by the health care provider before 29 weeks of gestation; being 5 km or less from the vaccination site; having two or more contacts with the health provider; and having some school education.

  Reasons reported by nonimmunized women for not receiving the vaccine included lack of awareness of the need for vaccination (44%), lack of information about the time and place of immunization session (29%) and fear of side effects (17%). Program-related reasons were long distance to the vaccination site (16%), inconvenient time of vaccination sessions (16%) and long waiting time (11%)." Gupta and Keyl 1998, Pg. 5

• 108
  • “3 doses of a TCV by 12 months of age and a booster dose given at some point in childhood or during pregnancy will provide immunity for mothers and their infants through the childbearing years.” GiveWell’s non-verbatim summary of a conversation with Dr. Neal Halsey, Professor of International Health and Director of the Institute for Vaccine Safety, Johns Hopkins Bloomberg School of Public Health, January 13, 2015, Pg. 1
  • “The antibody concentration and avidity and also the duration of protection depend on a number of factors, including the age of the vaccines and the number of and intervals between vaccine doses. Three DTP doses in infancy will give 3–5 years’ protection, a further dose or booster (e.g. in early childhood) will provide protection into adolescence, and 1 or 2 more booster(s) will induce immunity well through adulthood – a duration of 20–30 years has been suggested. Booster responses can still be elicited after intervals of 25–30 years, demonstrating the persistence of immunological memory.” WHO 2006, Pgs. 203-204
  • We’d guess that additional evidence on the effectiveness of different immunization schedules would come from studies comparing the level of antibodies in people receiving immunizations on different schedules: “A schematic picture of tetanus antitoxin response of adults following primary and booster immunization with tetanus toxoid is shown in Figure 2. The degree and duration of immunity increases with the number of tetanus toxoid doses given. One dose of tetanus toxoid ensures little, if any, protection. Two to four weeks after the second dose the mean level of tetanus antitoxin usually exceeds the minimum “protective” level of 0.01 IU / ml, although the percentage of poorly-protected persons can still be up to 10%. Immunity also declines with time. After one year the percentage of poorly-protected persons may increase to 20% and the mean titre may fall to the threshold level. A study in Papua New Guinea showed that 78% of women immunized during pregnancy with two 10 limits of flocculation (Lf) doses of adsorbed tetanus toxoid, had antitoxin levels above 0.01 IU/m1 for at least three years; the mean antitoxin level was about 0.03 IU/ml (Figure 3).” Borrow, Balmer and Roper 2006, Pg. 10

• 109

  "Whereas protection is incomplete after the first vaccine dose, protective concentrations of antitoxin are achieved in the majority of vaccines after completion of 2 doses; a third dose induces immunity in almost 100% of those immunized. The interval between the tetanus toxoid-containing doses should be at least 4 weeks. Longer intervals may increase the magnitude and duration of the immune response, but should not be a reason to delay immunization.” WHO 2006, Pg. 203

• 110

  "Concerns have been raised that maternal malaria could affect neonatal protection by reduction of maternal response to immunisation or placental antibody transfer. In some studies, malaria infection has been seen to decrease the antibody response to tetanus toxoid in children, although malaria chemoprophylaxis seems to preserve the response. A study comparing pregnant women with and without malaria parasitaemia noted no difference in the antibody response to tetanus toxoid; however, all participants received chloroquine prophylaxis. Studies investigating the effects of placental malaria on transplacental tetanus antibody transfer have had conflicting results: one showed reduced tetanus antibody concentrations in newborn babies of mothers with severe placental malaria, yet two others detected no such effect. Most studies of the immune response to tetanus toxoid in HIV-infected patients have been done in children and non-pregnant adults. Infants and adults infected with HIV generally do mount a protective response to tetanus toxoid, but their antibody levels tend to be lower than those of uninfected controls, especially in those whose CD4 lymphocyte counts are less than 300 cells per µL. With disease progression, immune response and serum tetanus antibody concentrations decrease. This blunted response suggests that HIV-infected individuals might need more frequent booster doses. Placental transfer of tetanus IgG was significantly lower in HIV-infected Brazilian mothers than in uninfected controls, a finding that was not replicated in a subsequent study in Malawi. In both studies, all neonates had protective tetanus antibody concentrations." Roper, Vandelaer, and Gasse 2007, Pgs. 1951-1952

• 111
  • See Treatment effect by doses (xlsx)
  • “Not a trial. 10-year follow up conducted on half of the area where Black 1980 was carried out.” Demicheli, Barale and Rivetti 2013, Pg. 19
  • "In July and August of 1974, the International Centre for Diarrhoeal Disease Research carried out a cholera vaccine trial in the entire Matlab area that included male and female children 1 to 14 years of age and non-pregnant women 15 years of age or older. Participants received, on a double-blind basis, either 1 or 2 doses of 0.5 mL of cholera toxoid or 0.5-mL adult-dose aluminum-phosphate-adsorbed tetanus-diphtheria toxoids as the placebo (with an interval of 42 days between the first and second injections).” Koenig et al. 1998, Pgs. 903-904
  • "In late 1977, the International Centre for Diarrhoeal Disease Research launched an experimental maternal and child health/family planning program in half of the Matlab study area, with the other half remaining as a comparison area. In the intervention area, maternal and child health services were gradually and carefully phased in over time; maternal tetanus immunization was one of the first interventions introduced.[12] Tetanus immunization coverage levels rose rapidly, reaching near universal levels by the early 1980s.” Koenig et al. 1998, Pg. 904
  • "Following the experimental design of the Matlab project, outreach health services in the comparison area were purposefully limited to those provided through the regular government program.[12] Vaccine trial participants from the comparison area were unlikely to have received additional tetanus immunizations prior to late 1988, when tetanus toxoid was introduced into the comparison area as part of the Bangladesh government's national immunization program. While no data exist on levels of maternal tetanus immunization coverage in the comparison area, these levels were very low prior to intensification of the national immunization program. Results from a 1989 national survey indicate that the maternal tetanus immunization coverage rate for rural Bangladesh as a whole remained at 5% or lower through 1984 and was still only 12% in 1986, rising significantly only toward the end of the decade.[13] Other evidence indicates that levels of health service coverage in the Matlab comparison area during this period are unlikely to have exceeded those for Bangladesh as a whole.” Koenig et al. 1998, Pg. 904