Oral Ivermectin Mass Drug Administration for Scabies Control

Summary

Published: September 2021

What is the problem?

Scabies is a skin condition caused by a parasitic mite that burrows into the outer layer of skin, causing intense itching and a rash that can become infected. Scabies-related infections often result in skin sores called impetigo. Crusted scabies is a more severe, much less common, and highly infectious form that involves many more mites than typical scabies.

Scabies infection is common in low- and middle-income countries. Prevalence varies widely but is highest in the Pacific and South America. According to Global Burden of Disease data, scabies imposes a similar total global health burden as tetanus and poisoning.

Scabies background

Scabies is a skin condition caused by a parasitic mite that burrows into the outer layer of skin, causing intense itching and a rash that can become infected. Crusted scabies is a more severe, much less common, and highly infectious form that involves many more mites than typical scabies.

According to the U.S. Centers for Disease Control and Prevention, scabies is a skin condition caused by the parasitic mite Sarcoptes scabiei var. hominis.1 Mites burrow into the outer layer of skin, laying eggs and causing “intense itching and a pimple-like skin rash”.2

Typical cases of scabies involve a small number of mites.3 Crusted scabies is an uncommon but more severe form of the condition involving very large numbers of mites, which causes crusting of the skin.4 This form occurs primarily in “the elderly, persons who are immunocompromised, or persons who have conditions that prevent them from itching and/or scratching (spinal cord injury, paralysis, loss of sensation, mental debility).”5 Crusted scabies is more common among people with untreated HIV/AIDS.6 Crusted scabies is highly infectious, making these individuals foci of scabies transmission in their communities.7

Scabies is primarily transmitted by skin-to-skin contact between people, and occasionally, the bedding or clothing of those who are infected.8

Prevalence and global distribution

Reported prevalence of scabies is highest in the Pacific islands, aboriginal populations in Australia, and South America, where prevalence ranges from approximately 2 to 37 percent. Prevalence tends to be highest in children. We have not vetted these estimates, and we are uncertain of how accurately they represent the prevalence of scabies in larger population groups.

Estimates of scabies prevalence and global distribution are based on sporadic surveys.9 Reported prevalence ranges widely from 0.2 to 71 percent, with highest prevalence reported in the Pacific (Pacific islands and aboriginal populations in Australia) and Latin America.10 Most of the results of prevalence surveys in the Pacific cluster between approximately 17 and 37 percent, while most of those in South America cluster between approximately 2 and 36 percent.11 We have not vetted these estimates. Some of the studies underlying these figures may represent higher-prevalence subpopulations of the nations in which they were conducted, so we are uncertain about their representativeness for larger population groups.

Prevalence tends to be highest in children and declines with age.12

We were unable to find prevalence figures for crusted scabies, but the World Health Organization refers to it as “uncommon.”13

Burden

Global Burden of Disease (GBD) assumptions imply that scabies does not cause significant mortality. Rather, its burden is imposed by “slight, visible physical deformity,” itching, and pain.14 However, expert opinion suggests that it can have consequences beyond those modeled by GBD, such as sleep deprivation, social stigma and isolation, absence from school or employment, and secondary bacterial infections resulting in skin sores (impetigo), heart damage, kidney damage, sepsis, and death.15 We have not identified estimates of the prevalence of these additional consequences, aside from impetigo, which ivermectin trials suggest is a common consequence of scabies infection.16

The 2015 GBD study assigned scabies a disability weight of 0.027, which implies that having scabies is equivalent in burden to having moderate hearing loss (0.027) or a heavy infestation of hookworm (0.027).17 GBD data also imply that the total global burden of scabies is similar to tetanus and poisoning.18 We have not investigated these figures in detail and we view them as highly uncertain.

What is the program?

In ivermectin mass drug administration, health workers distribute the oral antiparasitic drug ivermectin to almost all members of a scabies-affected community with the intention of controlling scabies mites.19

Community members typically receive one or two doses of ivermectin at 150 to 400 micrograms per kilogram of body weight,20 most commonly 200 µg/kg (see table below).

Since ivermectin is not recommended for young children and pregnant women, they receive a topical anti-scabies cream, usually containing the insecticide permethrin.21

Some interventions involve the identification and intensive treatment of people with crusted scabies, and due to its apparent importance in community scabies control, we include this element of the intervention in our evaluation.22

Does the program have strong evidence of effectiveness?

We identified nine intervention trials, all of which report a reduction in the prevalence of scabies lesions at longest follow-up. The most informative four trials, all conducted in island settings, report average suppression of scabies lesions of 91 percent at one year, 93 percent at two years, and 87 percent at three years. One long-term follow-up suggests that suppression may persist for at least 15 years, although we remain uncertain about the generalizability of this finding because it has not been replicated. The trials reporting the largest and most durable suppression treated entire island populations, while non-island trials reported a substantial rebound in scabies prevalence after initial suppression.

Although trials to date do not have the design strengths of typical randomized controlled trials, the large effect size, simple and plausible mechanism of action, and lack of alternative explanations strongly suggest that the intervention is effective for reducing the prevalence of scabies in most contexts that have been studied. The same trials suggest that the intervention is effective for reducing the prevalence of impetigo, although the evidence is strongest for follow-up periods of more than one year. A recent meta-analysis suggests that the intervention has similar effectiveness against soil-transmitted helminths as the commonly-used deworming drug albendazole, although it appears less effective against hookworm.

In our cost-effectiveness analysis, we model the benefits of ivermectin mass drug administration as occurring through reductions in morbidity from scabies and impetigo and gains in adult income from suppression of soil-transmitted helminths among children receiving ivermectin.

We identified several possible limitations of scabies mass drug administration, including that it may be less effective in non-island settings, it may promote resistance of scabies mites to ivermectin, and it may be significantly more complex and costly than other mass drug administration programs such as deworming. Evidence in animal models suggests that ivermectin may cause birth defects, but we are uncertain of the relevance of this finding to humans, or its possible incidence.

List of ivermectin mass drug administration trials

We performed a systematic scientific literature search for intervention trials of oral ivermectin mass drug administration for scabies control.23 We only included trials that were conducted in community settings and administered oral ivermectin to the majority of the population in a defined area.24

We identified nine intervention trials that meet these criteria. They are summarized in the table below.

Trial	Subjects	Intervention	Follow-up length	Impact on scabies	Notes
Bockarie et al. 2000	27 out of 31 people in a village in rural Papua New Guinea.25 Baseline scabies prevalence of 85% among treated.26	One dose of 400 µg/kg ivermectin.27	1, 2, 3, 4, and 5 months.28	Prevalence of scabies lesions declined from 85% at baseline to a nadir of 7% at two months, and rebounded to 15% at 5 months.29	The prevalence of scabies lesions in an untreated control village did not decline over a one-month follow-up period.30
Heukelbach et al. 2004	525 out of 605 people in a fishing village in Brazil.31 Baseline scabies prevalence of 3.8%.32	Two doses of 200 µg/kg ivermectin, separated by 10 days, except for pregnant/breastfeeding women and children <5. Mebendazole or albendazole for children <5. Topical deltamethrin lotion once per day for three days for those with scabies in the household and not receiving ivermectin.33	1 and 9 months.34	Prevalence of scabies lesions declined from 3.8% at baseline to 1% at one month and 1.5% at nine months.35	Intervention reduced the prevalence of several other parasitic infections including soil-transmitted helminths.36
Lawrence et al. 2005 and Marks et al. 2015	915 children and adults, constituting nearly all of the population of five small islands in the Solomon Islands. 25% prevalence of scabies infection among children at baseline, 20% among adults.37	1-2 treatments of 160–250 μg/kg ivermectin two weeks apart. Children <15 kg and pregnant women received 5% permethrin cream instead. Seven adults with crusted scabies were treated twice with both ivermectin and permethrin.38	4, 8, 12, 17, 22, 26, 32 and 36 months (36 month time point represents one island only),39 and 15 years.40	In children, prevalence of scabies lesions declined from 25% at baseline to 5% at four months and <1% at 32 months.41 In adults, prevalence of scabies lesions declined from 20% at baseline to 0.8% at 32 months.42 At 15 years, prevalence of scabies in children and adults was 0.3%.43	No change in scabies prevalence was observed after 4 months on a control island that did not receive the intervention.44 The number of skin sores (40% to 22%) and prevalence of hematuria declined in the treated group.45
Mohammed et al. 2012	The entire population of the islands of Zanzibar, Tanzania46 (population 1.3 million in 2012 census;47 70-80% coverage per round).48 Baseline number of cases varied widely based on location and age group.49	One dose of ivermectin and albendazole per year for six years. Dose not specified.50	1, 2, 3, 4, and 5 years.51	Reported cases of scabies declined in all locations and age groups. Relative decline in reported cases between baseline and 5 years ranged from 68 to 98 percent.52	Co-administration of the antiparasitic drug albendazole makes it difficult to isolate the specific impact of ivermectin. Reported cases of soil-transmitted helminths declined by 90-98% between baseline and 5 years.53
Kearns et al. 2015	960 of 1,256 residents of a remote Aboriginal community in Australia.54 Baseline scabies prevalence of 4%.55	Two annual doses of 200 µg/kg ivermectin. Those who weighed less than 15 kg or were pregnant received topical 5% permethrin or 10% crotamiton cream.56	6, 12, and 18 months.57	The prevalence of scabies lesions was 4% at baseline, 1% at six months, 9% at 12 months. Following the second dose of ivermectin at 12 months, the prevalence fell to 2 percent at 18 months.58	The authors argue that the increase at 12 months was caused in part by a case of “presumptive crusted scabies” in the community.59 The paper is unclear about whether the intervention treated the entire population of an island.
Romani et al. 2015b and Romani et al. 2019a	716 residents of three islands in Fiji in the ivermectin group. The total sample size across all three treatment groups represented 85% of the population of those islands.60 Baseline scabies prevalence of 32%.61	One dose of 200 µg/kg ivermectin; those diagnosed with scabies at baseline received a second dose one week later. Children weighing less than 15 kg, pregnant/lactating women, and people with other contraindications received topical permethrin cream.62 In a second group, topical permethrin was offered only to those diagnosed with scabies.63 In a third group, all subjects received one dose of topical permethrin cream, and a second dose after 1-2 weeks if they were diagnosed with scabies at baseline.64	1 and 2 years.65	In the ivermectin group, the prevalence of scabies lesions declined from 32% at baseline, to 1.9% at one year, to 3.6% at two years.66	Scabies lesions declined in the other two intervention groups as well, to a lesser degree. The prevalence of impetigo in the ivermectin group declined from 25% at baseline, to 8% at one year, to 2.6% at two years.67 In the ivermectin group, 82% of people were reached for follow-up at one year, and 90% at two years.
Martin et al. 2018	2,269 residents of eight villages in Tanzania.68 >85% coverage.69 Baseline scabies prevalence of 4.4%.70	One dose of 150-200 µg/kg ivermectin (dose not precisely specified) per year for three years.71 Children 5 years received topical benzyl benzoate cream instead, and pregnant/breastfeeding women presumably did as well, although it is not mentioned.72	1, 2, and 3 years.73	The prevalence of scabies lesions declined from 4.4% at baseline, to 0.8% at one year, then rose to 2.5% at two years, and 2.9% at three years.74	The paper lacks detail on the study population and intervention.
Marks et al. 2019	1,291 residents of six communities in the Solomon Islands, representing 89-93% of the population of those communities.75 Baseline scabies prevalence of 9-12%.76	One dose of 200 µg/kg ivermectin; those diagnosed with scabies at baseline received a second dose one week later. Children weighing <15 kg and pregnant/lactating women received topical permethrin cream. A second group received the same as above, and also one dose of 30 mg/kg azithromycin (max 2 g).77	1 year.78	In the ivermectin group, the prevalence of scabies lesions declined from 12% at baseline to 1.0% at one year. In the ivermectin + azithromycin group, the prevalence of scabies lesions declined from 9.2% at baseline to 0.7% at one year.79	The prevalence of impetigo in the ivermectin group declined from 10% at baseline to 2.5% at one year. The prevalence of impetigo in the ivermectin + azithromycin group declined from 12% at baseline to 3.3% at one year.80 75% of people in the ivermectin group were reached for follow-up, and 93% in the ivermectin + azithromycin group.81
Romani et al. 2019b and Marks et al. 2020	26,188 residents of three major islands in the Solomon Islands, representing 99% of the estimated total population. Baseline scabies prevalence of 19%.82	Two doses of 200 µg/kg ivermectin, spaced 1-2 weeks apart. Children weighing <12.5 kg and pregnant/lactating women instead received topical 5% permethrin cream in two doses 1-2 weeks apart.83 Subjects also received one dose of 20 mg/kg azithromycin, (1 g max).84	1 year and 3 years.85	The prevalence of scabies lesions declined from 19% at baseline to 2.3% at one year, then rebounded to 4.7% at three years.86	The prevalence of impetigo declined from 25% at baseline to 6.4% at one year, then rebounded to 9.6% at three years.87 Interpretation of impetigo results is complicated by azithromycin. Baseline and endline data reflect different subsets of the treated population, possibly introducing bias.88

What types of trials did we find?

The trials we identified can be treated as non-randomized intervention trials, although two were technically randomized. They range in size from 31 to over one million people, with four of nine targeting entire island populations. The most common drug regimen was one to two doses of ivermectin at 200 micrograms per kilogram. Follow-up length ranged from one month to 15 years, with seven trials reporting one-year data and five reporting data from two year follow-up or longer. All trials reported the prevalence of scabies lesions, while four reported the prevalence of impetigo. We believe Lawrence et al. 2005, Romani et al. 2015, Marks et al. 2019, and Romani et al. 2019b are the most informative trials.

Seven trials did not have control groups and presented before/after comparisons, while two were technically cluster-randomized controlled trials.89 However, the latter two trials did not have enough clusters (units of randomization) for effective randomization90 and did not include no-treatment comparison groups, so we treat them as non-randomized trials and focus on before/after comparisons in them as well.

The populations targeted for ivermectin treatment ranged in size from 31 people to over one million people, with seven of nine trials falling between 605 and 26,188 people. Four of nine trials targeted entire island populations, three of the four in Melanesia. We believe the distinction between island trials and non-island trials may be important because island trials report greater suppression of scabies over follow-up periods of more than one year, perhaps because they are more isolated and offer fewer opportunities for reinfection.

The interventions included one to six doses of oral ivermectin at 150 to 400 micrograms per kilogram of body weight, with the most common regimen being one to two doses of ivermectin at 200 micrograms per kilogram. In all but one trial, pregnant women, breastfeeding women, and young children received topical antiparasitic cream instead of ivermectin. Four trials provided more intensive treatment to those with scabies, or crusted scabies, at baseline.

All but two trials included a group that received only ivermectin, unless contraindicated. Mohammed et al. 2012 administered the antiparasitic drug albendazole in addition to ivermectin, which complicates its interpretation since albendazole has anti-scabies activity.91 Romani et al. 2019b administered the antibiotic azithromycin in addition to ivermectin, but this does not complicate the interpretation of the scabies outcome because azithromycin does not appear to impact scabies.92 Azithromycin may, however, complicate the interpretation of the impetigo outcome of this trial, since impetigo is caused by bacteria that are susceptible to azithromycin.93

Follow-up length ranged from one month to 15 years, with seven trials reporting one-year data and five reporting data from two year follow-up or longer. All trials reported the prevalence of scabies lesions. The only other commonly-reported outcomes are the prevalence of impetigo (four trials) and the prevalence of other parasitic infections such as soil-transmitted helminths (two trials).

We believe Lawrence et al. 2005, Romani et al. 2015, Marks et al. 2019, and Romani et al. 2019b are the most informative trials due to their large size, high baseline prevalence of scabies, intervention coverage of 85 percent or greater, detailed descriptions of the intervention, direct determination of scabies prevalence at endline, and follow-up of at least one year.

How effective is ivermectin mass drug administration over one year?

Taking the average of the four most informative trials, at one year ivermectin mass drug administration reduced the prevalence of scabies by 91 percent. At one year, there was not a consistent distinction between the findings of island vs. non-island studies. We believe the large effect size, simple and plausible mechanism of action, and lack of alternative explanations strongly suggest that the intervention is effective for reducing the prevalence of scabies over one year in most contexts that have been studied. The intervention also reduces the prevalence of impetigo at one year, although the effect size is highly variable. Ivermectin appears to have similar effectiveness against common intestinal worms as the deworming drug albendazole, except against hookworm, where it is less effective.

Scabies

In all but one trial, the prevalence of scabies declined in the year following ivermectin treatment. The four trials with the strongest designs report 88-94 percent suppression of scabies prevalence at one year: Lawrence et al. 2005, Romani et al. 2015, Marks et al. 2019, and Romani et al. 2019b. Taking a simple average, these four trials report a one-year reduction in the prevalence of scabies lesions of 91 percent.94

At one year, there was not a consistent distinction between the findings of island vs. non-island studies. For example, Martin et al. 2018 administered one dose of ivermectin per year for three years to eight villages in Tanzania. At the one year follow-up, the prevalence of scabies lesions had declined by 82 percent relative to baseline.95

We believe the large effect size, simple and plausible mechanism of action, and lack of alternative explanations strongly suggest that the intervention is effective for reducing the prevalence of scabies over one year in most contexts that have been studied.

The most notable exception is Kearns et al. 2015, in which the prevalence of scabies lesions 12 months after ivermectin treatment was more than twice the baseline value. The authors attribute this to a case of untreated crusted scabies in the community.96 This finding emphasizes the importance of identifying and treating cases of crusted scabies for long-term community control.

Impetigo

Four trials reported the prevalence of impetigo or unspecified skin sores at baseline and one year following the intervention, and in all cases it was reduced by 45 to 75 percent. These are the same four trials we identified as having the strongest designs. Romani et al. 2019b reported the second-largest reduction in impetigo prevalence but this finding is difficult to interpret because its intervention also included the antibiotic azithromycin.97

We believe the small to large effect size, simple and plausible mechanism of action, and lack of alternative explanations suggest that the intervention is effective to varying degrees for reducing the prevalence of impetigo over one year, primarily in isolated island contexts.

Soil-transmitted helminths

Two trials reported the prevalence of soil-transmitted helminths at baseline and at 9 to 12 months following the intervention, and in both cases the prevalence of soil-transmitted helminth species was reduced by 17 to 98 percent. However, Mohammed et al. 2012 co-administered the antiparasitic drug albendazole, complicating the interpretation of its findings. In the remaining trial, Heukelbach et al. 2004, suppression of soil-transmitted helminth prevalence at nine months ranged from 17 to 95 percent, depending on species.98

Due to the limited evidence from these trials, we sought additional evidence on the effectiveness of ivermectin for the control of soil-transmitted helminths. We identified one recent network meta-analysis comparing single-dose ivermectin to single-dose albendazole, which is an effective and widely-used deworming drug.99

It reports that ten days to six weeks after treatment, ivermectin has a cure rate 38 to 111 percent as large as albendazole for three common species of soil-transmitted helminths, with the lowest relative effectiveness reported for hookworm and the highest for whipworm.100

Together, we believe this evidence strongly suggests that ivermectin is effective for reducing the prevalence of soil-transmitted helminths over one year.

How effective is ivermectin mass drug administration over more than one year?

Two-year, three-year, and 15-year results from three of the four strongest trials suggest that ivermectin mass drug administration can achieve multi-year suppression of scabies lesions when the intervention reaches entire island populations. However, these trials provide somewhat conflicting evidence on the durability of suppression, with one suggesting that scabies lesions begin to rebound after three years, and another suggesting that they do not rebound at all after 15 years. We believe the large effect size, simple and plausible mechanism of action, and lack of alternative explanations strongly suggest that the intervention is effective for reducing the prevalence of scabies over more than one year in most contexts that have been studied, although the effect is more durable in trials that treated entire island communities. Two trials report greater suppression of impetigo at two years than at one year, despite the fact that ivermectin was only administered at baseline.

Scabies

Five trials report follow-up periods of more than one year, and in each case ivermectin treatment led to a reduction in the prevalence of scabies lesions at longest follow-up relative to baseline. Among the four strongest studies we identified, two reported two-year outcomes: Lawrence et al. 2005 and Romani et al. 2015. Taking a simple average of outcomes at two years, these two trials report a reduction in the prevalence of scabies lesions of 93 percent.101 This is similar to the level of suppression achieved at one year. Notably, these trials both treated entire island populations.

Among the four strongest studies we identified, two reported three-year outcomes: Lawrence et al. 2005 and Romani et al. 2019b. Taking a simple average of outcomes at three years,102 these two trials report a reduction in the prevalence of scabies lesions of 87 percent.103 A follow-up study of the Romani et al. 2019b intervention reports a modest but significant rebound in the prevalence of scabies lesions between one and three years, from 2.3 to 4.7 percent.104

Marks et al. 2015 is a 15-year follow-up study of scabies lesions in the population treated in Lawrence et al. 2005. This population had not received a mass scabies control intervention since the original trial, although we do not know how routine medical care for scabies may have changed over the follow-up period.105 The results suggest that the prevalence of scabies lesions had not rebounded after 15 years and remained suppressed by 99 percent.106

Therefore, two-year, three-year, and 15-year results from three of the four strongest trials suggest that ivermectin mass drug administration can achieve multi-year suppression of scabies lesions when the intervention reaches entire island populations. However, these trials provide somewhat conflicting evidence on the durability of suppression, with one suggesting that scabies lesions begin to rebound after three years, and another suggesting that they do not rebound at all after 15 years. We are uncertain how to interpret this apparent discrepancy.

Among the two studies targeting non-island populations and reporting follow-up periods of more than one year, the findings are less impressive, with both studies showing a rebound in the prevalence of scabies lesions from an earlier nadir. For example, Martin et al. 2018 administered one dose of ivermectin per year for three years to eight villages in Tanzania. Although the intervention achieved 81 percent suppression at one year, after three years the prevalence of scabies lesions was only suppressed by 34 percent relative to baseline.107 This is despite the fact that the research group administered ivermectin annually throughout the study period, with greater than 85 percent population coverage each round.108

We believe the moderate to large effect size, simple and plausible mechanism of action, and lack of alternative explanations suggest that the intervention is effective for reducing the prevalence of scabies over more than one year in most contexts that have been studied. The effect appears to be more durable in isolated island populations, although we are not certain that this is the reason for the divergence in long-term outcomes between these two classes of trials.

Impetigo

Two trials report the prevalence of impetigo or unspecified skin sores at baseline and at more than one year following the intervention, and they report 45 and 90 percent suppression at longest follow-up (31 and 24 months).109 Both interventions report greater suppression of impetigo at two years than at one year, despite the fact that ivermectin was only administered at baseline. These are two of the trials we identified as having the strongest designs, and both interventions targeted the entire population of islands and reported strong suppression of scabies lesions.

We believe the moderate to large effect size, plausible mechanism of action, and lack of alternative explanations suggest that the intervention is effective for reducing the prevalence of impetigo over more than one year in isolated island settings.

Soil-transmitted helminths

Mohammed et al. 2012 is the only trial that reported the prevalence of soil-transmitted helminths at baseline and at more than one year after the intervention. At the five-year follow-up point, reported cases of soil-transmitted helminths had declined by 90 to 98 percent, depending on location and age group.110 However, Mohammed et al. 2012 co-administered the antiparasitic drug albendazole and included six annual rounds of intervention, complicating the interpretation of its findings.

As discussed above, a meta-analysis of unrelated trials reports that single-dose ivermectin is similar in effectiveness to the commonly-used deworming drug albendazole over short follow-up periods, although it appears less effective for hookworm.111 We are unaware of reasons why the outcomes of the two drugs would diverge over longer follow-up periods, so we take this as fairly strong evidence that ivermectin is similar in effectiveness to albendazole over follow-up periods of more than one year.

Other benefits of ivermectin mass drug administration

Review papers suggest that scabies can have consequences beyond itching, rash, and impetigo, such as sleep deprivation, social stigma and isolation, absence from school or employment, heart damage, kidney damage, sepsis, and death.112 We have not identified estimates of the prevalence of these additional consequences, and we remain highly uncertain about what their prevalence may be, so we do not incorporate them into our cost-effectiveness analysis.

In other contexts, our cost-effectiveness analyses assume that interventions that substantially improve early-life health yield developmental benefits that result in lifetime increases in adult income.113 Although this is a possible outcome of treating scabies infection, we view it as unlikely due to the modest health burden imposed by scabies. However, we have only considered this possibility superficially. We do assume that the deworming effect of ivermectin yields lifetime increases in adult income,114 because this effect of deworming is supported by empirical evidence.115

Potential limitations and negative/offsetting effects

We identified several possible limitations of ivermectin mass drug administration, including that it may be less effective in non-island settings, it may promote resistance of scabies mites to ivermectin, and it may be significantly more complex and costly than other mass drug administration programs such as deworming. Evidence in animal models suggests that ivermectin may cause birth defects, explaining why it is generally not given to pregnant women, but we are uncertain of the relevance of this finding to humans or its possible incidence.

Weaker effects in non-island settings

The trials with the most impressive findings were conducted in isolated island communities, while trials in less isolated settings report less effective long-term control of scabies. Although we are not confident that this distinction is meaningful, due to the small number of trials in each category, a plausible interpretation is that less isolated communities are more susceptible to reinfection by nearby untreated communities.

Ivermectin resistance

Some researchers have suggested that scabies mites have become more resistant to ivermectin over time, and they have documented patients with crusted scabies who do not respond to ivermectin therapy.116 This raises the possibility that large-scale mass drug administration with ivermectin may be less effective against scabies than predicted because mites will become increasingly resistant to the drug over time. This offers a possible explanation for the findings of Martin et al. 2018, which reports robust suppression of scabies lesions at one year followed by substantial rebound over the following two years, despite ongoing annual administration of ivermectin to more than 85 percent of the population.117

Intervention complexity

For successful long-term scabies control, interventions may have to identify and intensively treat cases of crusted scabies. The US Centers for Disease Control and Prevention does not consider one or two doses of ivermectin sufficient for eradicating scabies mites in people with crusted scabies.118 Unless diagnosed and treated individually, these highly infectious individuals may remain foci of scabies transmission in their communities after ivermectin mass drug administration.119 Identifying and treating these individuals presumably increases the complexity and cost of the intervention.

Ivermectin is not recommended for young children and pregnant women. If the intervention is intended to target an entire community, it must include topical medications for those who cannot take ivermectin, and this presumably increases the complexity and cost of the intervention relative to a similar intervention like deworming that only involves an oral drug. It may also decrease compliance because applying a topical cream requires more effort than taking a pill.

Adverse effects of oral ivermectin

According to manufacturer information approved by the U.S. Food and Drug Administration, the recommendation against ivermectin treatment for pregnant women is explained by the fact that ivermectin can cause birth defects in animal models.120 The same source indicates that there are “no adequate and well-controlled studies in pregnant women” supporting this adverse effect, so it remains fairly uncertain.121 Although the intervention we evaluate here excludes pregnant women from ivermectin treatment, it is likely that some pregnant women would receive ivermectin, either because they are unaware they are pregnant, or due to a miscommunication or misapplication of treatment guidelines. We have not attempted to quantify this effect or incorporate it into our cost-effectiveness model due to its uncertainty.

According to the same source, the recommendation against ivermectin treatment for young children is explained by the absence of evidence of safety and effectiveness in that age group, rather than by evidence that it has adverse effects.122

The same source also indicates that people with onchocerciasis (river blindness, a parasitic disease mostly found in sub-Saharan Africa) may experience adverse effects from ivermectin treatment, ranging from skin swelling to severe impacts on brain function and/or death. The document describes these latter effects as rare.123

We are not aware of other adverse effects of ivermectin treatment.

How cost-effective is the program?

We created a cost-effectiveness estimate of ivermectin mass drug administration, based on data from ivermectin trials and a potential funding opportunity via the World Scabies Elimination Program. According to our cost-effectiveness model, while cost-effectiveness is higher in island settings than non-island settings, our best guess is that, across both settings, the program is below the range of cost effectiveness of the opportunities that we expect to direct marginal donations to (about 10x cash or higher, as of 2021).124 We remain uncertain about several key parameters, particularly the baseline prevalence of scabies, the duration of scabies suppression, and the importance of the disability imposed by scabies infection. The large difference in cost-effectiveness between island and non-island settings is explained primarily by differences in the baseline prevalence of scabies and differences in the duration of scabies suppression.

Note that our cost-effectiveness analyses are simplified models that do not take into account a number of factors. There are limitations to this kind of cost-effectiveness analysis, and we believe that cost-effectiveness estimates such as these should not be taken literally, due to the significant uncertainty around them. We provide these estimates (a) for comparative purposes and (b) because working on them helps us ensure that we are thinking through as many of the relevant issues as possible.

Disability burden due to individual cases of scabies is relatively low, compared to conditions that lead to mortality. However, especially in island settings, prevalence of scabies can be high and ivermectin mass drug administration can decrease prevalence substantially and potentially over several years at a relatively low cost, leading to the potential for high cost-effectiveness.

A sketch of the cost-effectiveness model is below. Since the intervention may have different effects in island vs. non-island settings, we modeled them separately.

• Prevalence of scabies and impetigo: For island settings, we estimate the prevalence of scabies to be 25% and the prevalence of impetigo to be 29%. For non-island settings, we estimate the prevalence of scabies and impetigo to be 4%.
• Effect of ivermectin on scabies and impetigo: For island settings, we assume 92% scabies suppression, which represents the mean of one- and two-year scabies suppression figures from high-quality trials that primarily treated entire island populations.125 For non-island settings, we base our estimate of scabies suppression on Martin et al. 2018, which is the only non-island trial with follow-up beyond 18 months. It reports suppression of 82% at one year, 43% at two years, and 34% at three years.126 For both settings, we apply an internal validity adjustment of 80% to account for concerns about study quality.
• Duration of benefits: For island settings, we assume benefits last ten years. This assumption is based on a single follow-up study reporting that this level of scabies suppression persisted for at least 15 years on an island where the entire population was treated.127 We discount this figure by one-third due to the fact that it has not been replicated and alternative explanations are conceivable. For non-island settings, we assume suppression would be negligible beyond three years.
• Moral weight of scabies and impetigo: We estimate the moral weight of averting scabies infection and impetigo using the GBD disability weight for both conditions, expressed in years lost to disability (YLDs),128 and our standard moral weight for one YLD. This yields an estimate that averting scabies in one person for one year has 0.05% of the value of averting the death of a child under five, and averting impetigo has 0.01% of the value of averting the death of a child under five.129
• Additional benefits from soil-transmitted helminths: Our cost-effectiveness estimate also includes indirect benefits to adult income from the suppression of intestinal worms in childhood.130 This accounts for 1% of total benefit for island settings, but 36% in non-island settings. This difference is because the benefits from averting scabies and impetigo are much smaller in non-island settings, leading to proportionally larger benefits from deworming.
• Cost: Based on data provided by the World Scabies Elimination Program, we estimate a cost of $6.75 per person for both island and non-island settings.131

We remain uncertain about several key parameters that could substantially influence cost-effectiveness. Major uncertainties in our model include:

• Moral weights for scabies and impetigo: We are very uncertain about the scabies disability weight figure from the Global Burden of Disease that is a key input in our analysis. We believe it is difficult to compare the moral weight of averting an intense itch and cosmetic flaw to the moral weight of averting death or increasing income. The disability weight of scabies also does not include possible indirect effects of scabies, such as sleep loss, loss of schooling or labor income, organ failure, or death. We do not currently have enough information to model these indirect effects separately and we are uncertain of their importance.
• Baseline prevalence of scabies, impetigo, and soil-transmitted helminths: Our estimates of baseline scabies, impetigo, and soil-transmitted helminth prevalence are uncertain, particularly in non-island settings. None of our estimates are based on direct evidence from the settings that the World Scabies Elimination Program proposes to work in. We are particularly uncertain about the baseline prevalence of these conditions in non-island settings.132
• Duration of benefits: Our estimates of the duration of scabies and impetigo suppression are very uncertain. In island settings, our estimate relies on one long-term follow-up study that reports almost complete suppression over 15 years.133 We discount this by using a duration of ten years, due to general skepticism and a second island trial that reports a partial rebound in the prevalence of scabies lesions between one and three years.134 Another source of uncertainty is the possibility that this program, or a similar one, might be funded within this time window by another organization.135 We have not factored this possibility into our analysis. In non-island settings, evidence is even more limited and our estimates of both the duration and magnitude of benefits rely on one trial.
• Differences between interventions in trials and interventions by potential charities: The trials that produced the effectiveness data in our model used interventions that differ somewhat from the intervention that the World Scabies Elimination Program proposes to implement. Although the World Scabies Elimination Program proposes to implement two annual rounds of treatment,136 the trials we rely upon to estimate the effectiveness of ivermectin mass drug administration typically offered 1-2 doses of ivermectin, with 7-14 days between doses if there was a second dose. This discrepancy between the interventions that have been tested, and those that would receive charitable funds, adds to our uncertainty about our cost-effectiveness estimates.
• Adverse effects: We do not include potential adverse effects on developing fetuses in mothers who do not know they are pregnant. We are uncertain about this possible adverse effect, and we do not have enough information to model it.

Given the very limited evidence base for several key inputs, we are particularly uncertain about our cost-effectiveness analysis for non-island settings.

To assess the sensitivity of our island model to key assumptions about uncertain parameters, we modeled a range of inputs. We find that most combinations of inputs suggest cost-effectiveness roughly similar to what we have included in our main cost-effectiveness estimates, but some can yield cost-effectiveness within the range that we would consider funding.137 We performed the same sensitivity analysis for our non-island model. We find that cost-effectiveness remains below programs we would recommend funding across all reasonable combinations of key assumptions.

Organizations and room for more funding

We identified one organization that implements ivermectin mass drug administration for scabies control, the World Scabies Elimination Program.138 Professor Andrew Steer, Director of Infection and Immunity at the Murdoch Children’s Research Institute, which manages the World Scabies Elimination Program, indicated that there may be additional countries where the World Scabies Elimination Program could begin conducting ivermectin mass drug administration for scabies, including island and non-island settings. We estimate that room for more funding would be approximately $4.7 million in these countries over 5 years of implementation.139

In our conversation, Dr. Steer stated that he is not aware of other organizations implementing ivermectin mass drug administration programs for scabies control.140

Key questions for further investigation

Questions we would ask as part of further investigation include:

• What is our level of confidence in the disability weight of scabies used in the 2015 Global Burden of Disease study, which is a key input in our cost-effectiveness analysis?
• Can we find higher-quality estimates of the prevalence of scabies in the populations that are most likely to be targeted by this intervention?
• Are there non-island settings that would meet the conditions for cost-effectiveness? If so, this may have the potential to substantially increase room for more funding.
• How much could ivermectin resistance impact cost-effectiveness?
• Are there other charities that operate in this area?
• Would the World Scabies Elimination Program secure funding without our involvement?

Our process

We searched U.S. Centers for Disease Control and Prevention and World Health Organization websites for general information on scabies. We performed shallow Google Scholar and Google searches for review papers on scabies prevalence using the search terms “scabies prevalence review.” We examined Global Burden of Disease data on the global distribution and burden of scabies and its disability weight.

We performed a systematic PubMed search on March 25, 2020 for trials of oral ivermectin mass drug administration for scabies control using the search terms “scabies AND ivermectin” with the “humans” filter activated. This returned 384 results, all of which we scanned for relevance. We also searched the references of the papers we identified for additional studies. We included trials that were conducted in the community and administered oral ivermectin to the majority of the population in a defined area.

We performed a systematic PubMed search on April 4, 2020 for meta-analyses on oral ivermectin for soil-transmitted helminth control using the search terms “ivermectin mass drug administration soil-transmitted helminths”. This returned 20 results, all of which we scanned for relevance.

Sources

Document	Source
Ayoub et al. 2009	Source
Bockarie et al. 2000	Source
CDC, Scabies, 2010	Source (archive)
CDC, Scabies, Biology, 2010	Source (archive)
CDC, Scabies, Disease, 2010	Source (archive)
CDC, Scabies, Medications, 2019	Source (archive)
Clarke et al. 2019	Source
Engelman et al. 2019	Source
GiveWell, CEA, 2021	Source
GiveWell, Combination deworming report, 2018	Source
GiveWell, Ivermectin MDA for scabies CEA, 2020	Source
GiveWell, Ivermectin MDA for scabies supporting calculations, 2020	Source
GiveWell’s non-verbatim summary of a conversation with Professor Andrew Steer and Matthew Parnaby, February 11, 2020	Source
Global Burden of Disease study 2015 disability weights	Source (archive)
Haar et al. 2014	Source (archive)
Hay et al. 2012	Source
Heukelbach et al. 2004	Source (archive)
IHME, GBD Results tool, 2017 DALYs from tetanus, scabies, and poisonings	Source (archive)
Ivers et al. 2012	Source (archive)
Karimkhani et al. 2017	Source (archive)
Kearns et al. 2015	Source (archive)
Lawrence et al. 2005	Source (archive)
Marks et al. 2015	Source (archive)
Marks et al. 2019	Source (archive)
Marks et al. 2020	Source (archive)
Martin et al. 2018	Source (archive)
Merck, Stromectol (ivermectin), New drug application 50-742/S-022	Source (archive)
Mohammed et al. 2006	Source
Mohammed et al. 2012	Source (archive)
Mounsey et al. 2009	Source
Murdoch Children’s Research Institute, World Scabies Elimination Program	Source (archive)
Romani et al. 2015a	Source
Romani et al. 2015b	Source
Romani et al. 2019a	Source
Romani et al. 2019b	Source (archive)
Rosumeck et al. 2018	Source (archive)
Sullivan et al. 1997	Source
Tanzania National Bureau of Statistics, 2012 Population and Housing Census, 2014	Source (archive)
WHO, Guideline: Alternative mass drug administration regimens to eliminate lymphatic filariasis, 2017	Source (archive)
WHO, Scabies, 2020	Source (archive)
WHO, Soil-transmitted helminth fact sheet, 2020	Source (archive)

• 1

  “Human scabies is caused by an infestation of the skin by the human itch mite (Sarcoptes scabiei var. hominis).” CDC, Scabies, 2010.

• 2

  “The microscopic scabies mite burrows into the upper layer of the skin where it lives and lays its eggs. The most common symptoms of scabies are intense itching and a pimple-like skin rash.” CDC, Scabies, 2010.

• 3

  “Because mites are often few in number (only 10-15 mites per person), these burrows may be difficult to find.” CDC, Scabies, Disease, 2010.

• 4
  • “Immunosuppressed individuals, including people living with HIV/AIDS, may develop an uncommon manifestation called crusted (Norwegian) scabies. Crusted scabies is a hyper-infestation with thousands to millions of mites, producing widespread scale and crust, often without significant itching.” WHO, Scabies, 2020.
  • “Because they are infested with large numbers of mites (up to 2 million), persons with crusted scabies are very contagious.” CDC, Scabies, Disease, 2010.

• 5

  CDC, Scabies, Disease, 2010.

• 6

  “Immunosuppressed individuals, including people living with HIV/AIDS, may develop an uncommon manifestation called crusted (Norwegian) scabies.” WHO, Scabies, 2020.

• 7
  • “The risk of transmission increases with the level of infestations, with highest risk due to contact with individuals with crusted scabies. Transmission due to contact with infested personal items (e.g. clothes and bed linens) is unlikely with common scabies but may be important for individuals with crusted scabies.” WHO, Scabies, 2020.
  • See description of medical treatment for classic scabies vs. crusted scabies. CDC, Scabies, Medications, 2019.

• 8
  • “Transmission occurs primarily by the transfer of the impregnated females during person-to-person, skin-to-skin contact. Occasionally transmission may occur via fomites (e.g., bedding or clothing).” CDC, Scabies, Biology, 2010.
  • “The most common source of transmission is prolonged skin-to-skin contact with an infected individual (hand-holding, sexual contact, etc.). It takes c. 15–20 min of close contact for successful direct transmission, and for this reason scabies is also considered to be a sexually transmitted disease.” Hay et al. 2012, Pg. 314.

• 9

  Romani et al 2015 is the most recent systematic review we identified of studies reporting scabies prevalence worldwide. Romani et al. 2015a, Figure 2.

• 10

  “The prevalence of scabies ranged from 0·2% to 71·4%. All regions except for Europe and the Middle East included populations with a prevalence greater than 10%. Overall, scabies prevalence was highest in the Pacific and Latin American regions, and was substantially higher in children than in adolescents and adults.” Romani et al. 2015a, abstract.

• 11

  Based on visual inspection of Figure 4 of Romani et al. 2015a.

• 12
  • “Overall, scabies prevalence was highest in the Pacific and Latin American regions, and was substantially higher in children than in adolescents and adults.” Romani et al. 2015a, abstract.
  • Also see Romani et al. 2015a, Figure 5.

• 13

  “Immunosuppressed individuals, including people living with HIV/AIDS, may develop an uncommon manifestation called crusted (Norwegian) scabies. Crusted scabies is a hyper-infestation with thousands to millions of mites, producing widespread scale and crust, often without significant itching.” WHO, Scabies, 2020.

• 14
  • Global Burden of Disease study 2015 disability weights, Scabies.
  • “The GBD disability weight method is founded on the principle that no disability double-counting occurs. Thus, although the skin effects of scabies can cause more notable morbidity such as sleep deprivation, mental disorders, and renal problems than all other skin conditions, the disability weight for scabies assesses only the direct effect of skin infection and the GBD 2010 Skin Conditions Expert Group assigned it the severity of disfigurement level 1.” Karimkhani et al. 2017, Pg. 1253.

• 15
  • “The best documented of these is the result of secondary infection by group A streptococci. For some time, it has been recognized that infection with streptococci may result in the development of glomerulonephritis. This can be detected with screening tests in a varying proportion of those affected, mainly children. For instance, symptomatic acute glomerulonephritis was reported in 10% of children in a survey in northern Australia, but, in addition, 24% had microscopic haematuria. Thus, asymptomatic renal damage can also occur. The infection was closely linked to skin sores, and scabies was identified as the principal cause. Infection with streptococci can also occur in the absence of scabies. Acute post-streptococcal glomerulonephritis is different in the tropics, as the skin, rather than the pharynx, appears to be the main source of infection. It has also been noted that persistent proteinuria can be detected for up to 16 years after the initial infection in 13% of those with recognized post-streptococcal glomerulonephritis vs. 4% of controls in an area endemic for scabies-associated infection. The real possibility exists that the renal damage that occurs after a primary attack of scabies with secondary infection may persist for years afterwards, with the potential to cause long-term glomerular damage. A further study has shown that control of scabies with ivermectin is also associated with a significant reduction in both haematuria and isolation of streptococci from skin lesions.” Hay et al. 2012, Pg. 317.
  • “Thus, although the skin effects of scabies can cause more notable morbidity such as sleep deprivation, mental disorders, and renal problems than all other skin conditions, the disability weight for scabies assesses only the direct effect of skin infection and the GBD 2010 Skin Conditions Expert Group assigned it the severity of disfigurement level 1.” Karimkhani et al. 2017, Pg. 1253.
  • Engelman et al. 2019, Figure 3.

• 16

  See the table in the section “Does the program have strong evidence of effectiveness?” for a summary of the impact of scabies control on impetigo, a common skin lesion caused by bacterial infection.

• 17

  The DALY ranges for these three conditions overlap. See Global Burden of Disease study 2015 disability weights.

• 18

  GBD Results Tool search can be found here.

• 19
  • “We therefore initiated a pilot study to determine the impact of community-based treatment with ivermectin on the prevalence of scabies in a previously untreated village.” Bockarie et al. 2000, Pg. 128.
  • “Control was achieved by treating almost all residents of each island once or twice within 2 weeks with ivermectin (160–250 microg/kg), except for children who weighed less than 15 kg and pregnant women, for whom 5% permethrin cream was used.” Lawrence et al. 2005, abstract.

• 20
  • “Control was achieved by treating almost all residents of each island once or twice within 2 weeks with ivermectin (160–250 microg/kg), except for children who weighed less than 15 kg and pregnant women, for whom 5% permethrin cream was used.” Lawrence et al. 2005, abstract.
  • “All participants in the mass treatment program were closely examined for signs of scabies before they were treated with a single 400 µg/kg dose of ivermectin.” Bockarie et al. 2000, Pg. 128.

• 21
  • “Control was achieved by treating almost all residents of each island once or twice within 2 weeks with ivermectin (160–250 microg/kg), except for children who weighed less than 15 kg and pregnant women, for whom 5% permethrin cream was used.” Lawrence et al. 2005, abstract.
  • “All non-pregnant participants who weighed ≥ 15 kg were administered a single dose of ivermectin 200 μg/kg at baseline and at month 12. Those ineligible for ivermectin received either topical 5% permethrin or 10% crotamiton.” Kearns et al. 2015, Pg. 4.

• 22

  “Treatment for crusted scabies consisted of the administration of two doses of ivermectin 1 week apart and the administration of permethrin cream twice a week for 1 month, with follow-up at 1, 2, 3, and 12 months.” Romani et al. 2015a, Pg. 2308.

• 23

  We performed a systematic PubMed search on March 25, 2020 for trials of oral ivermectin mass drug administration for scabies control using the search terms “scabies AND ivermectin” with the “humans” filter activated. This returned 384 results, all of which we title-scanned for relevance. We reviewed the full text of all relevant trials. We also searched the references of the papers we identified for additional studies.

• 24
  • We did not include ivermectin trials conducted in nursing homes, such as Sullivan et al. 1997, because we believe they have limited external validity to typical community ivermectin mass drug administration interventions.
  • We did not include Haar et al. 2014 because it only treated 49 percent of residents eligible for treatment.
  • We did not include a Cochrane meta-analysis on ivermectin for scabies, Rosumeck et al. 2018, because it does not focus on mass drug administration and only includes trials that compared ivermectin against permethrin. Abstract: “We included randomized controlled trials that compared permethrin or ivermectin against each other for people with scabies of all ages and either sex.” Pg. 36: “Public health questions like mass drug interventions for prevention and treatment of scabies were not addressed in this review.”

• 25
  • "The study was conducted in two villages in Dreikikir district in the East Sepik province of Papua New Guinea. One of the villages, Warasikau, was shortly to be included in a mass treatment program for the control of lymphatic filariasis. The other, Mimbiok, which served as the control village, was not part of the treatment program." Bockarie et al. 2000, Pg. 128.
  • "All the 31 residents of Warasikau present on the night of the filariasis survey were examined for scabies and 27 (87%) were considered positive. They were all treated with the exception of four who were below the age of 5 years because ivermectin was not approved for use in their age group at the time." Bockarie et al. 2000, Pg. 128.

• 26

  Due to the wording of the paper, we are not certain that our interpretation of this figure is correct. The paper states that everyone with scabies was given ivermectin, implying a prevalence of 100% among those treated, while later it cites a prevalence figure of 85% among those treated. This may reflect different diagnostic methods, with the 100% figure possibly resulting from visual identification of scabies-like lesions and the 85% figure possibly resulting from microscopic identification of scabies mites from skin scrapings, but the paper does not state this explicitly. “All the 31 residents of Warasikau present on the night of the filariasis survey were examined for scabies and 27 (87%) were considered positive. They were all treated with the exception of four who were below the age of 5 years because ivermectin was not approved for use in their age group at the time. Ivermectin is also not recommended for pregnant women but all women of childbearing age, examined in this study, indicated that they were not pregnant. Of the 27 persons given ivermectin, 23 (85%) were shown to have scabies.” Bockarie et al. 2000, Pg. 128.

• 27

  "All participants in the mass treatment program were closely examined for signs of scabies before they were treated with a single 400 µg:kg dose of ivermectin." Bockarie et al. 2000, Pg. 128.

• 28

  "Skin examinations were performed at monthly intervals for the subsequent 5 months." Bockarie et al. 2000, Pg. 128.

• 29

  "The prevalence decreased to 44.4% 1 month after treatment. The lowest monthly prevalence was observed 2 months after treatment when only 2 (7%) of the subjects had clinical signs of scabies. These two individuals were the mothers of young untreated infected children. A few scabetic lesions reappeared in six people 3 months after treatment but these new lesions resolved over the following 1 or 2 months. Five months after treatment, only four (14.8%) of those treated had scabetic lesions." Bockarie et al. 2000, Pg. 128.

• 30

  "In contrast to these findings, there was no change in the prevalence of scabies in the untreated village of Mimbiok in December 1996 and January 1997. Out of 25 persons, the number with scabies in December 1996 and January 1997 were 13 (52%) and 15 (60%), respectively." Bockarie et al. 2000, Pg. 128.

• 31

  "An intervention was carried out in a traditional fishing village in north-east Brazil where the population of 605 is heavily affected by ectoparasites and enteroparasites...A total of 525 people out of a target population of 576 were treated at baseline." Heukelbach et al. 2004, Pg. 563.

• 32

  Heukelbach et al. 2004, Table 3, Pg. 567.

• 33

  "All members of these households were treated with ivermectin (200 μg/kg; Revectina, Solvay Farma, São Paulo, Brazil), provided there were no contraindications. The dose was repeated after 10 days. Contraindications for administration of ivermectin were: being younger than 5 years, weighing < 15 kg, being pregnant or breastfeeding, or having renal or hepatic disease. Mebendazole or albendazole were administered to children who were < 5 years...In cases of pediculosis or scabies, topical deltamethrin lo-tion (Deltacid, Solvay Farma, São Paulo, Brazil) was given to individuals for whom ivermectin was contraindicated if the individual or at least one family member had pediculosis or scabies. Participants were instructed to use the lotion for three consecutive nights on their head (in cases of pediculosis) or their body (in cases of scabies)." Heukelbach et al. 2004, Pg. 565.

• 34

  "Follow-up examinations were performed at 1 month and 9 months after treatment." Heukelbach et al. 2004, Pg. 563.

• 35

  Heukelbach et al. 2004, Table 3, Pg. 567.

• 36

  Heukelbach et al. 2004, Table 2 and Table 3, Pg. 566-567.

• 37
  • "We studied the people of five small densely-populated islands (Sulufou, Foueda, Nuileni, Funafou and Addagege) in the Lau Lagoon, Malaita Province, Solomon Islands. Our census in-cluded 1558 people, about half of whom were on the islands at any one time." Lawrence et al. 2005, Pg. 35.
  • "Five hundred and forty-one different children were seen in the course of the study." Lawrence et al. 2005, Pg. 36.
  • "The prevalence of scabies [among children] fell dramatically on all the islands after treatment (Fig. 2), from a mean of 25%, 66 of 267 (95% CI, 20–30), to a steady level of less than 1%, 2 of 305 (0.7%; 95% CI, 0.1–2.2), P < 0.001." Lawrence et al. 2005, Pg. 36.
  • "At the time of treatment about 20% (81 of 374) of the adults for whom findings were recorded had scabies." Lawrence et al. 2005, Pg. 37.

• 38

  "Orally administered ivermectin was given at a dose of 160–250 μg/kg, using 6 mg scored tablets (15–18 kg, 3 mg; 19–27 kg, 4.5 mg; 28–36 kg, 6 mg; 37–55 kg, 9 mg; 56–74 kg, 12 mg; 75–100 kg, 15 mg). Children who weighed less than 15 kg were treated topically with 5% permethrin cream as were women who, in response to queries from a health worker, volunteered the information that they were or might be pregnant...The seven adults with severe generalized scabies were treated twice with both ivermectin and permethrin." Lawrence et al. 2005, Pg. 36.

• 39

  Lawrence et al. 2005, Figure 2, Pg. 37.

• 40

  "The first attempt at MDA for scabies control using ivermectin was conducted between 1997 and 2000 in five communities in the Lau Lagoon in the Solomon Islands...We conducted a cross-sectional survey to establish the prevalence of scabies and impetigo 15 years after the cessation of the scabies control programme in these communities." Marks et al. 2015, Pg. 3.

• 41

  Lawrence et al. 2005, Figure 2, Pg. 37.

• 42

  "At the time of treatment about 20% (81 of 374) of the adults for whom findings were recorded had scabies. At the last review when the hands and arms of about 250 adults from all islands were examined only two, neither of whom had been treated, were diagnosed as having scabies." Lawrence et al. 2005, Pg. 37.

• 43

  "Scabies was diagnosed in one individual (0.26%, 95% CI 0.0–1.4%)." Marks et al. 2015, Pg. 5.

• 44

  "Controls: On Foueda, two pre-treatment observations were made, in July and November 1997. At the time of the first observation, 23% (14/61) of the children seen had scabies or possible scabies; 14 had healed scabies. The proportion of children with scabies and those with sores did not change significantly between July and November (26%; 12/46)." Lawrence et al. 2005, Pg. 37.

• 45
  • "The proportion of children with sores decreased from 40%, 91 of 226 (95% CI, 34–47) to 22%, 35 of 162 (95% Lawrence et al. 2005, Pg. 37.
  • "There was a significant decrease in the proportion of children with haematuria at both 25 cells/microlitre (P = 0.002) and 10 cells/microlitre (P < 0.001), despite wide variation from month to month." Lawrence et al. 2005, Pg. 38.

• 46

  "In Zanzibar, the Programme for Eliminating Lymphatic Filariasis (PELF) administered donated drugs, namely ivermectin (Mectizan) and albendazole, annually to all eligible individuals for six consecutive years (2001–05), through a community-based delivery system." Mohammed et al. 2012, Pg. 1.

• 47

  See the table on Pg. 2 of Tanzania National Bureau of Statistics, 2012 Population and Housing Census, 2014.

• 48

  "Drug distribution coverage for the total Zanzibar population averaged from 70 to 80% in allrounds, differing little from the coverage reported on tally sheets." Mohammed et al. 2006, Pg. 342.

• 49

  See Mohammed et al. 2012, Figure 1, Pg. 4.

• 50

  "In Zanzibar, the Programme for Eliminating Lymphatic Filariasis (PELF) administered donated drugs, namely ivermectin (Mectizan) and albendazole, annually to all eligible individuals for six consecutive years (2001–05), through a community-based delivery system." Mohammed et al. 2012, Pg. 1.

• 51

  "To determine the impact of MDA for LF on STH and scabies, 50 primary health care units (PHCU) in the ten districts of Zanzibar were chosen. Five PHCUs were selected at random in each district and the record registers were used to obtain data of registered cases of STH and scabies during a six year period (2000–2005)." Mohammed et al. 2012, Pg. 2.

• 52

  "Results: Health centre records showed a consistent decline in the number of cases of intestinal helminths and scabies diagnosed by community health workers in Zanzibar and the number of prescriptions issued across five age groups. A 90-98% decline in soil transmitted helminths and 68-98% decline in scabies infections were recorded." Mohammed et al. 2012, Pg. 1.

• 53

  "A 90-98% decline in soil transmitted helminths and 68-98% decline in scabies infections were recorded." Mohammed et al. 2012, Pg. 1.

• 54
  • "The setting was a remote island community, 550km from Darwin, Australia with an estimated population of 2121." Kearns et al. 2015, Pg. 3.
  • "At baseline, there were 1256 residents on the household occupancy lists in the population census (March-September 2010), of which 1013 (81%) consented to participate. Most participants (n = 960, 95%) were seen over a four month period (April-July)." Kearns et al. 2015, Pg. 3

• 55

  "Scabies prevalence among the baseline cohort was 4% and remained relatively stable during the initial assessment period." Kearns et al. 2015, Pg. 5.

• 56

  "All non-pregnant participants who weighed ≥ 15 kg were administered a single dose of ivermectin 200μg/kg at baseline and at month 12. Those ineligible for ivermectin received either topical 5% permethrin or 10% crotamiton." Kearns et al. 2015, Pg. 4.

• 57

  See Figure 1 of Kearns et al. 2015, Pg. 6.

• 58

  See Figure 1 of Kearns et al. 2015, Pg. 6.

• 59

  "The rise in scabies prevalence at month 12 coincided with: a cluster of cases epidemiologically-linked to an individual with presumptive crusted scabies, a high prevalence amongst new entries to the cohort (an indicator of the impact of high population mobility), and an increased prevalence amongst members of the baseline cohort who did not have a known exposure to the suspected crusted scabies case (4% to 8%)." Kearns et al. 2015, Pg. 10.

• 60
  • "We conducted SHIFT in Fiji from September 2012 through September 2013. After consultation with health authorities, we identified three island communities as study groups." Romani et al. 2015b, Pg. 2306.
  • "In the ivermectin group, 623 of the 716 participants received ivermectin under the direct observation of study staff, and 93 received permethrin, in accordance with the exceptions specified in the protocol." Romani et al. 2015b, Pg. 2309.
  • Also see Figure 2 of Romani et al. 2015b, Pg. 2310.

• 61

  See Table 2 of Romani et al. 2015b, Pg. 2310.

• 62

  "All participants in the ivermectin group were offered one dose of oral ivermectin (200 μg per kilogram of body weight), which was taken under the direct observation of study staff. Ivermectin was replaced with topical permethrin cream in the following participants: children who weighed less than 15 kg, women who were pregnant or breast-feeding, persons with neurologic disease, and persons taking medications that are metabolized by the cytochrome P-450 pathway, including warfarin and some anticonvulsant agents. For participants who had scabies at baseline, a second dose of the medication that had been provided at baseline was distributed by study staff 7 to 14 days after the initial dose was administered." Romani et al. 2015b, Pg. 2307-2308.

• 63

  "Participants in the standard-care group who had scabies at baseline were referred to the local clinic for guideline-recommended treatment with one dose of topical permethrin cream, with a second dose provided after 14 days if symptoms persisted." Romani et al. 2015b, Pg. 2307.

• 64

  "All participants in the permethrin group were offered one dose of topical permethrin cream followed by a second dose 7 to 14 days later if scabies had been observed at baseline. Romani et al. 2015b, Pg. 2307.

• 65
  • "The primary outcome was the change in the prevalence of scabies and of impetigo from baseline to 12 months. Prevalence was calculated at baseline and 12 months with the use of data from the entire sample of participants at each time point." Romani et al. 2015b, Pg. 2,308.
  • "In September 2015, we performed a follow-up survey of scabies and impetigo 24 months after the original mass drug administration in the communities enrolled in SHIFT to determine the durability of the effect of a single round of treatment." Romani et al. 2019a, Pg. 1.

• 66
  • "At 12 months, the ivermectin-based regimen was the most effective, with a 94% relative reduction from baseline in the prevalence of scabies (from 32.1% at baseline to 1.9%)." Romani et al. 2019a, Pg. 1.
  • "The prevalence of scabies at 24 months was 3.6% in the ivermectin group." Romani et al. 2019a, Pg. 2.

• 67

  See Figure 1 of Romani et al. 2019a, Pg. 1.

• 68
  • "We evaluated the impact of annual ivermectin MDA given for LF, following World Health Organization (WHO) guidelines, on the community-wide prevalence of scabies in eight treatment-naive villages in Kongwa district, Tanzania." Martin et al. 2018, Pg. 937.
  • Also see Table 1, Martin et al. 2018, Pg. 937.

• 69

  "Mass drug administration coverage was > 85% each year." Martin et al. 2018, Pg. 937.

• 70

  "Scabies rates of all ages dropped from 4.4% at baseline to 0.84% in Year 2 and then increased to 2.5% in Year 3 and 2.9% in Year 4 (Table 2)." Martin et al. 2018, Pg. 938.

• 71
  • "We evaluated the impact of annual ivermectin MDA given for LF, following World Health Organization (WHO) guidelines, on the community-wide prevalence of scabies in eight treatment-naive villages in Kongwa district, Tanzania." Martin et al. 2018, Pg. 937.
  • "Box 1. Existing WHO recommendations for MDA to eliminate lymphatic filariasis...2. Annual MDA of ivermectin (150–200 µg/kg) with albendazole (400 mg) in communities where LF
    and onchocerciasis are co-endemic." WHO, Guideline: Alternative mass drug administration regimens to eliminate lymphatic filariasis, 2017, Pg. 3.

• 72

  "In our study, a high proportion of scabies cases were younger than 5 years, who were offered benzyl benzoate cream rather than ivermectin, and this treatment was not directly observed." Martin et al. 2018, Pg. 939.

• 73

  "Baseline data collection occurred from October to December 2012 with follow-up data collection in October–December each year until 2015." Martin et al. 2018, Pg. 937

• 74

  See Table 2, Martin et al. 2018, Pg. 938.

• 75
  • "This was a community randomised open label study conducted in Malaita province of the Solomon Islands. Six communities were randomized to one of 2 arms: an ivermectin arm or a combined-treatment arm." Marks et al. 2019, Pg. 928.
  • Also see Table 1 of Marks et al. 2019, Pg. 929.

• 76

  "At baseline prevalence of scabies was 11.8% (95% confidence interval [CI] 9.4–14.6%) in the ivermectin-only arm and 9.2% (95% CI 7.1–11.7%) in the combined-treatment arm." Marks et al. 2019, Pg. 929.

• 77

  "In the ivermectin arm we administered ivermectin MDA at baseline. In the combined treatment arm we co-administered ivermectin and azithromycin MDA at baseline. Ivermectin MDA consisted of a single oral dose of ivermectin (200 μg/kg) determined by body weight. In individuals with a contra-indication to ivermectin (pregnancy, breast-feeding, weight <15 kg) topical permethrin was offered instead. Individuals clinically diagnosed with scabies at baseline were offered a second dose of ivermectin (or second application of topical permethrin) at 7 days. Azithromycin MDA consisted of a single oral dose of azithromycin (30 mg/kg, max 2 gm) determined by body weight." Marks et al. 2019, Pg. 928.

• 78

  "At baseline and 12 months, participants underwent a standardized examination by an experienced clinician (MM) with data recorded on the presence or absence of any skin lesions, their location, and whether they were consistent with scabies, impetigo, or another diagnosis." Marks et al. 2019, Pg. 928.

• 79

  "At baseline the prevalence of scabies was 11.8% (95% con-fidence interval [CI] 9.4–14.6%) in the ivermectin-only arm and 9.2% (95% CI 7.1–11.7%) in the combined-treatment arm...At 12 months the prevalence of scabies and impetigo had fallen to 1.0% (95% CI 0.3–2.6%) and 2.5% (95% CI 1.4–4.5%), respectively, in the ivermectin-only treatment arm and to 0.7% (95% CI 0.2–1.8%) and 3.3% (95% CI 2.1–5.1%), respectively, in the combined treatment arms (Table 2)." Marks et al. 2019, Pg. 929.

• 80

  "At baseline the prevalence of active impetigo was 10.1% (95% CI 8.1–13.0%) in the ivermectin-only treatment arm and 12.1% (95% CI 9.7–14.9%) in the combined-treatment arm...At 12 months the prevalence of scabies and impetigo had fallen to 1.0% (95% CI 0.3–2.6%) and 2.5% (95% CI 1.4–4.5%), respectively, in the ivermectin-only treatment arm and to 0.7% (95% CI 0.2–1.8%) and 3.3% (95% CI 2.1–5.1%), respectively, in the combined treatment arms (Table 2)." Marks et al. 2019, Pg. 929.

• 81

  See Marks et al. 2019, Table 1, Pg. 929.

• 82

  "During September, 2015, over 4 weeks, 26188 people (99·3% of the estimated population of Choiseul [n=26372] as determined at the 2009 census) were treated. At baseline, 1399 (84·2%) of 1662 people living in the first ten villages had their skin examined, of whom 261 (18·7%) had scabies and 347 (24·8%) had impetigo." Romani et al. 2019b, Pg. 510.

• 83

  "For ivermectin-­based mass drug administration, participants were offered either two doses of oral ivermectin 200 μg/kg 7–14 days apart using weight­-based bands, or if ivermectin was contraindicated (such as for pregnant and breastfeeding women and children weighing less than 12·5 kg) they were instead offered two applications of topical 5% permethrin cream 7–14 days apart." Romani et al. 2019b, Pg. 512-513.

• 84

  "Participants were offered a single dose of 20 mg/kg azithromycin, up to a maximum of 1 g, using weight-­based bands." Romani et al. 2019b, Pg. 512.

• 85
  • "The main outcome of the study (a secondary outcome in the original AIM trial) was the prevalence of scabies and impetigo in the ten randomly selected villages at 12 months compared with ten different randomly selected villages at baseline." Romani et al. 2019b, Pg. 513.
  • "We conducted a prevalence survey for scabies and impetigo in 10 villages in Choiseul Province of the Solomon Islands 36 months after a single round of ivermectin and azithromycin mass drug coadministration." Marks et al. 2020, Pg. 1591.

• 86

  See Table 2 of Marks et al. 2020, Pg. 1593.

• 87

  See Table 2 of Marks et al. 2020, Pg. 1593.

• 88

  "The main outcome of the study (a secondary outcome in the original AIM trial) was the prevalence of scabies and impetigo in the ten randomly selected villages at 12 months compared with ten different randomly selected villages at baseline. We chose a new set of villages at 12 months to ensure that our estimates of prevalence were as generalisable as possible to the whole province." Romani et al. 2019b, Pg. 513.

• 89
  • “We randomly assigned three island communities to one of three different interventions for scabies control: standard care involving the administration of permethrin to affected persons and their contacts (standard-care group), mass administration of permethrin (permethrin group), or mass administration of ivermectin (ivermectin group).” Romani et al. 2015b, abstract.
  • “Six communities were randomized to receive either ivermectin-based MDA or ivermectin-based MDA co-administered with azithromycin.” Marks et al. 2019, abstract.

• 90

  Marks et al. 2019 and Romani et al. 2018 only have one “cluster” per treatment group. Ivers et al. 2012 discusses the problem with this. “Given a large enough sample, ‘simple’ (or complete) randomization is expected to produce balance in C-RCTs across cluster-level covariates because this is the level of allocation. Unfortunately, many C-RCTs do not have enough clusters to create a reasonable expectation for cluster-level balance. Consider the review of C-RCTs published between 2000 and 2008, which found that the median number of clusters was 21, but 25% had fewer than 12 and 14% had less than four clusters per arm (the fewest recommended to ensure statistical validity).” Ivers et al. 2012, Pg. 2.

• 91
  • “A decrease in the number of STH and scabies cases diagnosed by health centre staff from 50 health facilities in Unguja and Pemba after the LF elimination programme had distributed ivermectin and albendazole between 2000–2005 was found in all age groups.” Mohammed et al. 2012, Pg. 4.
  • “Our observations represent, to our knowledge, the first cases of scabies successfully treated with oral albendazole and suggest that this agent provides a potential therapeutic option for scabies.” Ayoub et al. 2009, Pg. 175.

• 92

  Marks et al. 2019 report that ivermectin alone has the same impact on scabies prevalence as ivermectin plus azithromycin. This is expected because azithromycin is an antibiotic and therefore should not impact scabies mites. “Co-administration of azithromycin with ivermectin led to similar decreases in scabies and impetigo prevalence compared to ivermectin alone.” Marks et al. 2019, abstract.

• 93
  • Although Marks et al. 2019 report that ivermectin alone has the same impact on impetigo prevalence as ivermectin plus azithromycin, we remain cautious in our interpretation of this finding since the bacterial species responsible for impetigo, Staphylococcus aureus and Streptococcus pyogenes, are susceptible to azithromycin.
  • “Co-administration of azithromycin with ivermectin led to similar decreases in scabies and impetigo prevalence compared to ivermectin alone.” Marks et al. 2019, abstract.
  • “The macrolide azithromycin is a potential candidate for this role, because it has good activity against S. pyogenes and S. aureus.” Marks et al. 2019, Pg. 927.

• 94

  The calculations for this can be found here.

• 95

  “Scabies rates of all ages dropped from 4.4% at baseline to 0.84% in Year 2 and then increased to 2.5% in Year 3 and 2.9% in Year 4 (Table 2).” Martin et al. 2018, Pg. 938.

• 96

  “In addition to the new cohort entries, the increased prevalence at month 12 was influenced by a cluster of cases epidemiologically linked to a participant diagnosed with presumptive crusted scabies. Prevalence within the baseline cohort of those who were known contacts rose from 7% (7/96) at baseline to 18% (17/96) at month 12, whereas prevalence amongst others who were not known contacts within the baseline cohort rose from 4% (23/598) at baseline to 8% (46/604) at month 12.” Kearns et al. 2015, Pg. 5.

• 97
  • “The azithromycin ivermectin mass drug administration (AIM) trial was a prospective, single-arm, before-and-after, community intervention study to assess the efficacy of mass drug administration of ivermectin for scabies and impetigo, with coadministration of azithromycin for trachoma.” Romani et al. 2019a, abstract.
  • Although Marks et al. 2019 report that ivermectin alone has the same impact on impetigo prevalence as ivermectin plus azithromycin, we remain cautious in our interpretation of this finding since the bacterial species responsible for impetigo, Staphylococcus aureus and Streptococcus pyogenes, are susceptible to azithromycin.
  • “Co-administration of azithromycin with ivermectin led to similar decreases in scabies and impetigo prevalence compared to ivermectin alone.” Marks et al. 2019, abstract.
  • “The macrolide azithromycin is a potential candidate for this role, because it has good activity against S. pyogenes and S. aureus.” Marks et al. 2019, Pg. 927.

• 98

  Heukelbach et al. 2004, table 2.

• 99
  • Clarke et al. 2019.
  • “The WHO recommended medicines –albendazole (400 mg) and mebendazole (500 mg) – are effective, inexpensive and easy to administer by non-medical personnel (e.g. teachers). They have been through extensive safety testing and have been used in millions of people with few and minor side-effects.” WHO, Soil-transmitted helminth fact sheet, 2020.

• 100

  The paper reports a relative risk of cure for single-dose ivermectin relative to single-dose albendazole of 1.01 for Ascaris, 0.377 for hookworm, and 1.112 for whipworm. Clarke et al. 2019, Figures 3, 5, and 7.

• 101

  The calculations for this can be found here.

• 102

  We used 32-month data from Lawrence et al. 2005. This is because the 36-month follow-up data only represent one of five intervention islands and are therefore not comparable to earlier time points. “Sulufou only” is marked above the bar representing scabies prevalence at 36 months in figure 2. Lawrence et al. 2005, Figure 2, Pg. 37.

• 103

  The calculations for this can be found here.

• 104

  The 95 percent confidence intervals for these two values do not overlap, suggesting that the apparent rebound in the prevalence of scabies lesions is unlikely to be due to chance. “2.3% (1.6 to 3.3); 4.7% (3.6 to 6.1)”. Marks et al. 2020, Table 2, Pg. 1593.

• 105

  “Since the cessation of the study in 2000 no further active control measures have been undertaken in these communities.” Marks et al. 2015, Pg. 3.

• 106
  • “Three hundred and thirty eight residents were examined, representing 69% of the total population of the five communities. Only 1 case of scabies was found, in an adult who had recently returned from the mainland.” Marks et al. 2015, abstract.
  • See our calculations of percent suppression in this spreadsheet. The calculations for this can be found here.

• 107

  “Scabies rates of all ages dropped from 4.4% at baseline to 0.84% in Year 2 and then increased to 2.5% in Year 3 and 2.9% in Year 4 (Table 2).” Martin et al. 2018, Pg. 938.

• 108

  “Mass drug administration coverage was > 85% each year.” Martin et al. 2018, Pg. 937.

• 109
  • “The proportion of children with sores decreased from 40%, 91 of 226 (95% CI, 34–47) to 22%, 35 of 162 (95% CI, 16–28; P < 0.001) and the mean number of sores per affected child decreased with time (Table 1).” Lawrence et al. 2005, Pg. 37.
  • “The relative reduction from baseline in the prevalence of impetigo was greatest in the ivermectin group at 90% (95% CI, 74 to 99), as compared with a relative reduction of 64% (95% CI, 42 to 84) in the permethrin group and 39% (95% CI, 15 to 61) in the standard-care group.” Romani et al. 2015b, Pg. 2.

• 110

  Mohammed et al. 2012, Tables 1 and 3.

• 111

  The paper reports a relative risk of cure for single-dose ivermectin relative to single-dose albendazole of 1.01 for Ascaris, 0.377 for hookworm, and 1.112 for whipworm. Clarke et al. 2019, Figures 3, 5, and 7.

• 112
  • “The best documented of these is the result of secondary infection by group A streptococci. For some time, it has been recognized that infection with streptococci may result in the development of glomerulonephritis. This can be detected with screening tests in a varying proportion of those affected, mainly children. For instance, symptomatic acute glomerulonephritis was reported in 10% of children in a survey in northern Australia, but, in addition, 24% had microscopic haematuria. Thus, asymptomatic renal damage can also occur. The infection was closely linked to skin sores, and scabies was identified as the principal cause. Infection with streptococci can also occur in the absence of scabies. Acute post-streptococcal glomerulonephritis is different in the tropics, as the skin, rather than the pharynx, appears to be the main source of infection. It has also been noted that persistent proteinuria can be detected for up to 16 years after the initial infection in 13% of those with recognized post-streptococcal glomerulonephritis vs. 4% of controls in an area endemic for scabies-associated infection. The real possibility exists that the renal damage that occurs after a primary attack of scabies with secondary infection may persist for years afterwards, with the potential to cause long-term glomerular damage. A further study has shown that control of scabies with ivermectin is also associated with a significant reduction in both haematuria and isolation of streptococci from skin lesions.” Hay et al. 2012, Pg. 317.
  • “Thus, although the skin effects of scabies can cause more notable morbidity such as sleep deprivation, mental disorders, and renal problems than all other skin conditions, the disability weight for scabies assesses only the direct effect of skin infection and the GBD 2010 Skin Conditions Expert Group assigned it the severity of disfigurement level 1.” Karimkhani et al. 2017, Pg. 1253.
  • Engelman et al. 2019, Figure 3.

• 113

  For example, see our cost-effectiveness analysis for vitamin A supplementation.

• 114

  See our cost-effectiveness analysis of ivermectin mass drug administration for scabies.

• 115

  See our cost-effectiveness analysis of deworming here.

• 116

  “Survival times of mites exposed to ivermectin increased from 1997 through 2006 (P less than .001) (Table), while survival times remained unchanged over 10 years for the negative control emulsifying ointment. Even when we excluded the 2 patients with previously documented resistance2 from the analysis, we found that the trend for increasing ivermectin survival time remained significant (P = .006) (Table). Results from a single patient with recurrent CS and previously documented ivermectin resistance are of note (Figure). When this patient was treated again in 2006 with 3 doses of ivermectin (200 μg/kg), no noticeable reduction in mite numbers was observed (unpublished observations). A significant increase in mite survival time was observed when mites collected after 8 days and 3 doses of ivermectin were compared with those collected prior to the commencement of ivermectin therapy (P = .003).” Mounsey et al. 2009, abstract.

• 117
  • “Scabies rates of all ages dropped from 4.4% at baseline to 0.84% in Year 2 and then increased to 2.5% in Year 3 and 2.9% in Year 4 (Table 2).” Martin et al. 2018, Pg. 938.
  • “Mass drug administration coverage was > 85% each year.” Martin et al. 2018, Pg. 937.

• 118

  “Crusted scabies: both oral and topical agents should be used… Depending on infection severity, ivermectin should be taken in three doses (approximately days 1, 2, and 8), five doses (approximately days 1, 2, 8, 9, and 15), or seven doses (approximately days 1, 2, 8, 9, 15, 22, and 29)... Topical permethrin should be administered every 2-3 days for 1-2 weeks to treat crusted scabies… A topical keratolytic cream may also be used to help reduce the crusting of the skin and aid in the absorption of the topical permethrin or benzyl benzoate.” CDC, Scabies, Medications, 2019.

• 119

  Kearns et al. 2015 reports a large rebound in scabies prevalence following ivermectin mass drug administration, linked to a case of untreated presumptive crusted scabies. “In addition to the new cohort entries, the increased prevalence at month 12 was influenced by a cluster of cases epidemiologically linked to a participant diagnosed with presumptive crusted scabies. Prevalence within the baseline cohort of those who were known contacts rose from 7% (7/96) at baseline to 18% (17/96) at month 12, whereas prevalence amongst others who were not known contacts within the baseline cohort rose from 4% (23/598) at baseline to 8% (46/604) at month 12.” Kearns et al. 2015, Pg. 5.

• 120

  “Ivermectin has been shown to be teratogenic in mice, rats, and rabbits when given in repeated doses of 0.2, 8.1, and 4.5 times the maximum recommended human dose, respectively (on a mg/m2/day basis). Teratogenicity was characterized in the three species tested by cleft palate; clubbed forepaws were additionally observed in rabbits. These developmental effects were found only at or near doses that were maternotoxic to the pregnant female. Therefore, ivermectin does not appear to be selectively fetotoxic to the developing fetus.” Merck, Stromectol (ivermectin), New drug application 50-742/S-022, Pg. 6.

• 121

  “There are, however, no adequate and well-controlled studies in pregnant women. Ivermectin should not be used during pregnancy since safety in pregnancy has not been established.” Merck, Stromectol (ivermectin), New drug application 50-742/S-022, Pg. 6.

• 122

  “Safety and effectiveness in pediatric patients weighing less than 15 kg have not been established.” Merck, Stromectol (ivermectin), New drug application 50-742/S-022, Pg. 6.

• 123

  “After treatment with microfilaricidal drugs, patients with hyperreactive onchodermatitis (sowda) may be more likely than others to experience severe adverse reactions, especially edema and aggravation of onchodermatitis. Rarely, patients with onchocerciasis who are also heavily infected with Loa loa may develop a serious or even fatal encephalopathy either spontaneously or following treatment with an effective microfilaricide. In these patients, the following adverse experiences have also been reported: pain (including neck and back pain), red eye, conjunctival hemorrhage, dyspnea, urinary and/or fecal incontinence, difficulty in standing/walking, mental status changes, confusion, lethargy, stupor, seizures, or coma. This syndrome has been seen very rarely following the use of ivermectin. In individuals who warrant treatment with ivermectin for any reason and have had significant exposure to Loa loa-endemic areas of West or Central Africa, pretreatment assessment for loiasis and careful post-treatment follow-up should be implemented.” Merck, Stromectol (ivermectin), New drug application 50-742/S-022, Pg. 5.
  

• 124
  • As of 2021, we expect the marginal dollar we recommend will be about 10 or more times as cost-effective as GiveDirectly's unconditional cash transfer program. For an example of the cost-effectiveness of our recommendations, see this page.
  • See our cost-effectiveness analysis of oral ivermectin MDA for scabies control, Sheet “CEA,” Row “Cost-effectiveness in multiples of cash transfers.”

• 125

  Our calculations for this figure can be found here.

• 126

  Our calculations for these figures can be found here.

• 127

  “We found an extremely low prevalence of scabies 15 years after the cessation of a scabies control programme. The prevalence of impetigo had also declined further since the end of the control programme.” Marks et al. 2015.

• 128

  Global Burden of Disease study 2015 disability weights.

• 129

  Calculations:

  • For scabies, we use a disability weight of 0.027, assume 1 YLD equals 2.3 units of value and assume averting the death of a child under five has 116.9 units of value. Then the value for scabies is 0.027 x 2.3 / 116.9 = 0.053% the value of averting the death of a child under five.
  • For impetigo, we use a disability weight of 0.006, assume 1 YLD equals 2.3 units of value and assume averting the death of a child under five has 116.9 units of value. Then the value for scabies is 0.006 x 2.3 / 116.9 = 0.012% the value of averting the death of a child under five.
  • Moral weights underlying these calculations can be found here. GiveWell's 2020 moral weights.

• 130

  For more on these effects, see our page on deworming.

• 131

  For cost data, we rely on an estimate from the World Scabies Elimination Program that the total cost of the intervention will be between $1.25 and $1.45 USD per person per year over five years: “The World Scabies Program expects the cost of its ivermectin MDA programs to be between US$1.25 - $1.45 per person per year for two treatment rounds over the five years of the program. This cost per person per year is averaged over the five years, but there are variations in costs year to year. The first four years are more intensive and include costs for program establishment, surveillance set-up and support, medication purchase and shipping, capacity building, MDA delivery over two years, and follow-up. The final year is to ensure sustainability and to evaluate the impact of the program.” GiveWell’s non-verbatim summary of a conversation with Professor Andrew Steer and Matthew Parnaby, February 11, 2020.
  

• 132

  For island settings, our estimates are based on high-quality direct evidence from the Solomon Islands and Fiji, and we extrapolate those estimates to other islands in Oceania that we do not have direct evidence on. This extrapolation is based on personal communications from members of the World Scabies Elimination Program suggesting that the prevalence of scabies is similar in the two contexts, but we have not seen convincing evidence to support this assessment. For non-island settings, we rely on Global Burden of Disease Data that we are highly uncertain about. One source of our uncertainty is that the data represent entire nations while ivermectin mass drug administration might be targeted to higher-prevalence subsets of a country. We adjusted for this in our estimate by increasing baseline prevalence twofold, but this is highly speculative. We are unable to share the locations in consideration at this time.

• 133

  “Three hundred and thirty eight residents were examined, representing 69% of the total population of the five communities. Only 1 case of scabies was found, in an adult who had recently returned from the mainland.” Marks et al. 2015, abstract.

• 134

  Marks et al. 2020 reports a prevalence of scabies of 2.3 percent at one year and 4.7 percent at three years. Marks et al. 2020, Table 2, Pg. 1593.

• 135

  We also considered applying time discounting because scabies prevalence may decline over time due to growing medical infrastructure and affluence over time, but we ultimately omitted it because historical data suggests that the burden of scabies has not declined significantly between 1990 and 2017 in the Pacific islands. See the data underlying this judgment here.

• 136

  “The World Scabies Program expects the cost of its ivermectin MDA programs to be between US$1.25 - $1.45 per person per year for two treatment rounds over the five years of the program. This cost per person per year is averaged over the five years, but there are variations in costs year to year. The first four years are more intensive and include costs for program establishment, surveillance set-up and support, medication purchase and shipping, capacity building, MDA delivery over two years, and follow-up. The final year is to ensure sustainability and to evaluate the impact of the program.” GiveWell’s non-verbatim summary of a conversation with Professor Andrew Steer and Matthew Parnaby, February 11, 2020.

• 137

  When we did this sensitivity analysis, we found the following:

  • If we halve the baseline prevalence of scabies and impetigo, cost-effectiveness is below the range of programs we would recommend funding. Based on prevalence surveys, we believe it is unlikely that baseline prevalence would be substantially greater than what we assume so we did not model larger values.
  • If we assume that the benefits of scabies and impetigo suppression last three years rather than ten years, cost-effectiveness is below the range of programs we would recommend funding. If we assume that the benefits last 15 years, as implied by Marks et al. 2015, cost-effectiveness is within the range of programs we would recommend funding.
  • If we assume that the disability weight of scabies is overestimated twofold, cost-effectiveness is below the range of programs we would recommend funding. If we assume that it is underestimated twofold, for example because it does not incorporate indirect harms of scabies, cost-effectiveness is within the range of programs we would recommend funding.
  • Combining two or more of these changes can lead to fourfold or greater changes in cost-effectiveness, yielding estimates that are well below the range of programs we would recommend funding, or above it.

• 138
  • Murdoch Children’s Research Institute, World Scabies Elimination Program.
  • “In August 2019, MCRI and its partners were awarded AU$10 million by the Macquarie Group Foundation to create the World Scabies Program, which is supporting implementation of ivermectin-based MDA for scabies control in Fiji and the Solomon Islands over five years, reaching a total population of approximately 1.5 million.” GiveWell’s non-verbatim summary of a conversation with Professor Andrew Steer and Matthew Parnaby, February 11, 2020.

• 139

  During our conversation with the World Scabies Elimination Program, they mentioned two countries that had expressed interest with a combined population of 415,000. Given costs of $6.75 per person ($1.35 per person over 5 years), this suggests a total funding need of $2.8 million (equals $6.75 x 415,000). World Scabies Elimination Program has asked us to keep the names of these countries confidential. In a later communication, they mentioned an additional $1.9 million funding opportunity.

• 140

  “Professor Steer is not aware of other organizations operating ivermectin-based MDA programs.” GiveWell’s non-verbatim summary of a conversation with Professor Andrew Steer and Matthew Parnaby, February 11, 2020.