The Spatial-Temporal Distribution of Chronic Lymphatic Filariasis in Indonesia: A 18-Year Registry-Based Analysis

Round 1

Reviewer 1 Report

This is a very nicely written and comprehensive presentation of LF clinical cases in Indonesia over the past 18 years. It is important that this information be presented and documented in the peer reviewed literature and I would like to start by applauding Indonesia for their ongoing efforts in compiling and reporting these data.   I have only minor concerns, which I will make clear below:

Line 106: says that the definition is "people who are infected by parasitic helminths and show chronic symptoms..." does this mean that a positive mf or antigen test is required to be a positive diagnosis? If so, these clinical case counts would be a drastic undercounting of the true LF morbidity because it has been shown that chronic cases rarely have active infection.  Please clarify how cases are defined.

Methods section: In general, can you clarify if these are incident or cumulative cases? I presumed cumulative (meaning if a case is found in 2003, they are still reported in 2005). Is that correct? It would be good to be explicit in how cases are counted. 

Methods section: can you provide greater detail on how case finding takes place? Is it ongoing or does it coincide with MDA? Is it likely to be the same across provinces or might the degree of active case finding play a role in the differing LF case rates?

Figures: Please add titles for the y-axes in each figure; it will help with interpretation

Line 158: missing "one" in between "than" and "case". 

Line 159: need a "that" in between "provinces" and "recorded"

Figures 2 & 3: I find it very shocking that the absolute number of cases shown in Figure 2 so closely resembles the same relative pattern as the prevalence rate shown in Figure 3 across all the provinces.  That suggests to me that each province has a relatively similar underlying populations - is this true? The biggest exception to this is in West Papua where the population must be significantly smaller and thus the prevalence rate is much higher relative to other provinces. No need to edit anything if the data are correct - I am just noting that I would have expected a greater change in the relative rates by province as compared to the number of cases.

Line 192: Should be 'cases'

Line 202: "Micro filariasis" should be one word

Line 205: I don't follow the logic that a decrease in MDA coverage would explain a significant increase in LF cases that same year (or even the following year). LF morbidity takes years and years to accumulate. On the other hand, if MDA in Aceh generally has lower coverage, then as the population ages, I may expect to see less of a reduction in incident LF cases than in other provinces with high MDA coverage (which aligns with the India modelling paper referenced). 

Line 212: replace 'trebles' with 'triples'

Line 217: Can you update 2020 to the current year?

Line 219: missing a 'but' before 'rather'

Author Response

Thank you for your very constructive comments. We have addressed them point-by-point as seen below. We believe that the peer-review process has now resulted to a much-strengthened manuscript.

- Line 106: says that the definition is "people who are infected by parasitic helminths and show chronic symptoms..." does this mean that a positive mf or antigen test is required to be a positive diagnosis? If so, these clinical case counts would be a drastic undercounting of the true LF morbidity because it has been shown that chronic cases rarely have active infection.  Please clarify how cases are defined.

Indeed, you are correct that this would be an underestimate. We have added considerably more detail in the methodology, explaining the National protocol that is being followed.

Thus, we have added the following in the methodology:

“According to the Regulation of the Ministry of Health: No 94 (2014), diagnostic tests are performed to detect the presence of microfilaria by microscopic investigation, and IgG4 antifilarial antibody detection by rapid test and ELISA. Finger Prick Bloods (FPB) are collected and prepared into a thick-blood smear slide and subsequently stained using Giemsa or Wright staining before being examined under the microscope. The rapid test measurement uses recombinant antigens (BmR1 and BmSXP) and the results are characterized by the development of 2-3 strips (bands) indicating positive results or the presence anti-filarial antibody of IgG4 in sample serum. The instrument used for detecting the presence of Brugia spp. infection is the Brugia Rapid test; while for detecting W. Bancrofti, the ICT bancrofti or Pan LF kits are utilized. In some circumstances, the measurement of IgG4 antifilarial antibody level is done using ELISA technique, though the application of ELISA remains scarce within the dataset. As per the above regulation, individuals who live in endemic areas and present stage 1 symptoms (i.e., swelling in the leg that usually disappears when they wake up in the morning) will be tested using the LF rapid test. Individuals who are in stages 2-3 will have rapid test or blood examination to confirm the stage of the disease, and individuals in stages 4-7 will be provided with follow-up clinical examination(s).”

And in the discussion section

“As the data was collected from within the province, initiated at the community level, we consider that this study can capture true and representative population-level data, though it should be noted that clinical case counts are likely to be a drastic under-counting of the true LF morbidity, as it has been shown that chronic cases rarely have active infection”

- Methods section: In general, can you clarify if these are incident or cumulative cases? I presumed cumulative (meaning if a case is found in 2003, they are still reported in 2005). Is that correct? It would be good to be explicit in how cases are counted.

Thank you for your comment. Indeed, these are cumulative cases. We have added the below in the methodology section.

“The data presented in the current manuscript therefore consists of confirmed, cumulative LF cases at both district and province level across the entire area of Indonesia as reported between January 2001 and December 2018.”

- Methods section: can you provide greater detail on how case finding takes place? Is it ongoing or does it coincide with MDA? Is it likely to be the same across provinces or might the degree of active case finding play a role in the differing LF case rates?

Thank you for raising this point. The case finding is continuous, driven by primary-health centres and community nurses. The MDA has strengthened the intensity of the monitoring in general but has not changed its continuous nature.

Having said that, your point is exactly correct, in that healthcare coverage is not proportionally the same across the many islands of Indonesia, and this might play a role in the differing LF case rates.

We have made the following additions to the text:

“The data used in the current manuscript was collected from the smallest reporting units within Indonesia, i.e., reported by the village/district, through case finding from the head of village and/or health care staff. Case finding is carried on by those units on a continuous basis.”

And in the limitations section

“Furthermore, some cases may be reported twice as there is a small background of internal migration within Indonesia, and the records cannot identify such cases. In addition, the continuous monitoring of cases, is heavily dependent on the smaller healthcare units, however their relative proportional population coverage differs between islands, and this might play a role in the differing LF cases reported.”

- Figures: Please add titles for the y-axes in each figure; it will help with interpretation

Thank you, it has been addressed as indicated

- Line 158: missing "one" in between "than" and "case".

Thank you, it has been addressed as indicated

- Line 159: need a "that" in between "provinces" and "recorded"

Thank you, it has been addressed as indicated

- Figures 2 & 3: I find it very shocking that the absolute number of cases shown in Figure 2 so closely resembles the same relative pattern as the prevalence rate shown in Figure 3 across all the provinces.  That suggests to me that each province has a relatively similar underlying populations - is this true? The biggest exception to this is in West Papua where the population must be significantly smaller and thus the prevalence rate is much higher relative to other provinces. No need to edit anything if the data are correct - I am just noting that I would have expected a greater change in the relative rates by province as compared to the number of cases.

Indeed, you are correct in your observation. The West Papuan population is significantly smaller than on any of the other major islands, this resulting to the observed prevalence pattern. Additionally, West Papua is mostly covered by densely forested areas and is more resource-constrained than Indonesia as a whole, worsening the overall risk of contracting vector-borne diseases.

- Line 192: Should be 'cases'

Thank you, it has been addressed as indicated

- Line 202: "Micro filariasis" should be one word

Thank you, it has been addressed as indicated

- Line 205: I don't follow the logic that a decrease in MDA coverage would explain a significant increase in LF cases that same year (or even the following year). LF morbidity takes years and years to accumulate. On the other hand, if MDA in Aceh generally has lower coverage, then as the population ages, I may expect to see less of a reduction in incident LF cases than in other provinces with high MDA coverage (which aligns with the India modelling paper referenced).

Thank you for your comment, and apologies if it was not clear. MDA in Aceh has a much lower coverage than in Java Island. This is a resource-driven decision by the Indonesian Ministry of Health, as roughly 55% of Indonesia’s total population resides on Java Island. We have added the following for clarity in the discussion section.

“The relationship between coverage and LF cases was explained in India by the mathematical modelling of lymphatic filariasis elimination programs, showing that in high endemicity such as Aceh and West Nusa Tenggara 4-12 rounds are needed with a minimum coverage 50% of the total area. In the case of Aceh, where MDA generally has a lower coverage, as the population ages, one would expect to see less of a reduction in incident LF cases than in other provinces with high MDA coverage.”

- Line 212: replace 'trebles' with 'triples'

Thank you, it has been addressed as indicated.

- Line 217: Can you update 2020 to the current year?

Thank you, it has been addressed as indicated. Indeed, this is correct, it was originally meant to be evaluated in 2020, but due to the pandemic the evaluation will take place in the current year (2022).

- Line 219: missing a 'but' before 'rather'

Thank you, it has been addressed as indicated.

Reviewer 2 Report

This is an interesting study and covers diverse aspects of epidemiology in the study area.

Author Response

Thank you very much for your support and constructive comments.

We have addressed some of the minor aspects that were requested through the peer-review process, providing further details for enhanced clarity, resulting to a much-strengthened manuscript.