The claim that Gaza’s health ministry has historically reported accurate mortality data rests on one sentence in a one-page correspondence letter published in The Lancet on 6 December 2023, in which Huynh, Chin and Spiegel put the discrepancy against other analyses at 1.5% to 3.8% in previous conflicts and supply no reference marker. Their own measurement covers 35 days of 2023 and nothing earlier, and the quoted endpoints reproduce exactly, and only on the Gaza Ministry of Health’s own count as the denominator, from three number pairs an Associated Press explainer of 26 October 2023 records without computing a percentage. A search of the biomedical and preprint literature and of the open web on 10 September 2026 located no peer-reviewed study comparing a pre-2023 ministry count against any other body’s count.
Key Findings
- The historical-accuracy claim traces to one sentence in a one-page correspondence letter published online 6 December 2023, which states a discrepancy of 1.5% to 3.8% in previous conflicts and, verified in the full text and again in the typeset PDF, carries no reference marker (Huynh, Chin and Spiegel, 2024).
- The letter’s own measurement is a 35-day comparison running 7 October to 10 November 2023, in which Huynh, Chin and Spiegel report 5.3 deaths per 1,000 against 7.8 per 1,000 on two different populations. They measure nothing before 2023.
- Computed from the three number pairs the Associated Press explainer of 26 October 2023 records, 4 of 260 gives 1.5% and 55 of 1,440 gives 3.8%, both on the Gaza Ministry of Health’s own count as the base; on the counterpart base the same figures give 1.6% and 4.0% and appear nowhere.
- The Gaza Ministry of Health states 28,185 records with complete data and 9,715 whose data was not yet complete as of 30 June 2024, and defines the pair itself. Both are that ministry’s own count, not independently verified.
- 5,817 of those 28,185 records carry the source label of notification by the martyrs’ relatives, against 22,368 labelled ministry records, on the provenance column the Gaza Ministry of Health publishes.
Where does the claim of historical accuracy come from?
The claim that the Gaza Ministry of Health has historically reported accurate mortality data traces, in every place it circulates, to one sentence in a one-page correspondence letter by Huynh, Chin and Spiegel, published online in The Lancet on 6 December 2023. The sentence is quoted here whole, because the sentence is the evidence.
The Lancet published it under the title “No evidence of inflated mortality reporting from the Gaza Ministry of Health,” in an issue dated 6 January 2024, under DOI 10.1016/S0140-6736(23)02713-7. It is Correspondence rather than a research article, so it is editorially reviewed and not peer-reviewed. That status is disclosed here at first use and travels with every figure taken from it.
The sentence reads: “The Gaza MoH has historically reported accurate mortality data, with discrepancies between MoH reporting and independent United Nations analyses ranging from 1·5% to 3·8% in previous conflicts.” The journal sets decimals with a middle dot, so the printed strings are 1·5% and 3·8%; this report writes 1.5% and 3.8% in its own prose.
The title is itself an absence claim, and it is routinely upgraded in circulation. “No evidence of inflated” is not “verified,” not “accurate,” and not “confirmed,” and Huynh, Chin and Spiegel state in the same letter that mortality “might have been under-reported by the Gaza MoH due to decreased capacity.” They add that difficulties obtaining accurate mortality figures “should not be interpreted as intentionally misreported data,” which is a caution against an inference in one direction rather than a finding in the other (Huynh, Chin and Spiegel, 2024).
What did that one-page letter actually measure?
35 days, and nothing before 2023. The letter’s own measurement is a temporal comparison running from 7 October to 10 November 2023, in which Huynh, Chin and Spiegel set the Gaza Ministry of Health’s reported deaths against reported staff deaths published by the United Nations Relief and Works Agency over the same window.
The two published figures are, in the letter’s words, “Cumulative reported deaths were 101 UNRWA staff members and 11 078 Gazans over 35 days.” From those, Huynh, Chin and Spiegel derive two rates, 5.3 deaths per 1,000 against 7.8 per 1,000, and the inference they draw is narrow: the second rate exceeding the first is inconsistent with inflation of the first.
Those two rates run over two different populations, Gaza residents and one agency’s Gaza staff, so both denominators travel together or neither does. The letter’s data statement scopes its own data to October and November 2023: no pre-2023 comparison was conducted and no method for one is stated (Huynh, Chin and Spiegel, 2024).
The letter also relays a peacetime baseline it did not measure, an average of 4,884 registered deaths a year in Gaza across 2015 to 2019, which the Palestinian Central Bureau of Statistics produced. That is civil registration in peacetime and the 11,078 is wartime reporting over five weeks, so the two are never divided into one another.
What the letter measured is a five-week rate comparison inside the current war. What the quoted sentence asserts is a record of accuracy across earlier wars, and only the first was tested.
Does the 1.5% to 3.8% range carry a citation?
The historical-accuracy sentence carries zero reference markers, verified in the journal’s HTML full text and again in the typeset PDF on 10 September 2026. Markers 1 and 2 close the sentence before it, about political scepticism, and marker 2 closes the sentence after it, about an 8.0% figure, so the range stands in a journal as a bare assertion.
The letter is short enough that this is easy to see rather than easy to miss. Huynh, Chin and Spiegel number their references in the ordinary way, and the markers sit on the sentences bracketing the range on either side.
A reference marker is the mechanism by which a journal converts an author’s statement into a checkable one. Without it, the strongest available reading is that Huynh, Chin and Spiegel state the range on their own authority, in a piece the journal classes as Correspondence rather than as refereed research.
This matters out of proportion to its size because of where the sentence travels. The range is the most widely quoted casualty-method figure on this subject, and every restatement of it passes through this one uncited line. A reader who follows the citation back looking for a study finds a letter, and inside the letter finds no pointer at all for that clause.
The range is nonetheless reproducible, because the letter carries a reference two sentences later, and the source behind that marker records numbers from which both endpoints recover exactly. That is arithmetic rather than retrieval, and it is set out in full below (Huynh, Chin and Spiegel, 2024).
Which three wars do those percentages reproduce from?
Three dated number pairs, and no percentage at all, sit in the Associated Press explainer of 26 October 2023, updated 6 November 2023, that the letter points to two sentences later. It records the Gaza Ministry of Health’s own counts for three earlier conflicts, 1,440 for 2008, 2,310 for 2014 and 260 for 2021, each beside a counterpart toll, and computes nothing.
Computed here on 10 September 2026, a difference of 4 on the 2021 pair gives 1.5% of 260, and one of 55 on the 2008 pair gives 3.8% of 1,440. Those are the two published endpoints, recovered exactly. Each difference is the gap between the two tolls inside a pair, and each divides into the Gaza Ministry of Health’s figure rather than into its counterpart.
The 2014 pair falls at 2.55%, and that value appears nowhere in the letter. It is therefore the interior of the range rather than an endpoint, which is what a range built from three values of this kind would look like.
The November 2012 conflict is absent from the explainer altogether, so “previous conflicts” in the quoted sentence means three of them rather than every earlier war.
The counterpart figures in the pairs are post-war tolls published by the United Nations Office for the Coordination of Humanitarian Affairs, named here only as what the Associated Press explainer of 26 October 2023 records. They carry no finding in this report. The answering figures are the Gaza Ministry of Health’s own counts of 1,440, 2,310 and 260, and the arithmetic reproduced from them above.
Whose count is the denominator in that comparison?
The Gaza Ministry of Health’s own count, and only its own. Every percentage the letter publishes reproduces exactly when the ministry’s figure is the denominator, and none of them reproduces on the counterpart figure, so the range is a share of a party’s own count rather than of a separately produced one. Huynh, Chin and Spiegel never state what the range is a share of.
On the ministry base the three differences give 1.5%, 3.8% and 8.0%, matching the letter exactly. On the counterpart base the same three differences give 1.6%, 4.0% and 8.7%, and none of those values appears anywhere in the letter (Huynh, Chin and Spiegel, 2024).
The third value is the letter’s own separate figure. Its next sentence states that a comparison between the Gaza health ministry and Israeli Foreign Ministry mortality figures for the 2014 war “yielded an 8·0% discrepancy,” and the Israeli Foreign Ministry reported 2,125 in that comparison, that ministry’s own count, published by a party to the conflict and not independently verified.
The word doing the heaviest work in the quoted sentence is “independent.” The Associated Press explainer the letter points to does not use it. Its strongest phrasing is that the ministry’s counts “have held up to U.N. scrutiny,” and the same explainer records that the humanitarian office worked from medical records, which in Gaza are the ministry’s own. On that relay the two counts are not established as drawn from separate record systems, so the adjective belongs to the letter rather than to the source it cites.
Has any study compared a pre-2023 count with another body’s?
No such study was located in a search run on 10 September 2026, across the biomedical and preprint paper index in two framings returning 30 results, and across the open web. Nothing in the published literature compares a pre-2023 Gaza Ministry of Health count against any other body’s count of the same deaths. The absence is recorded here, and no claim is made in either direction.
What a study of that kind would have to do is specific. It would need two counts of the same deaths, over the same window, produced through record systems that do not feed one another, with a stated matching rule and a stated treatment of the records each system misses.
The nearest items the search returned are each something else. Vaktskjold and colleagues, publishing in Conflict and Health in 2016, ran a chart review of non-war mortality in Gaza for July to September 2014, recording 1,241 deaths against 1,232 in the corresponding period of 2013 (Vaktskjold and colleagues, 2016). It is a peer-reviewed measurement of mortality, and it runs no count-against-count comparison.
The only peer-reviewed test ever turned on the ministry’s registry is the capture-recapture analysis Jamaluddine and colleagues published in The Lancet in 2025, and it covers 7 October 2023 to 30 June 2024. It is a test of the current war and says nothing about any earlier one.
The record therefore holds a widely quoted assertion about three earlier conflicts, number pairs from which it reproduces arithmetically, and no study that ever ran the comparison.
What does the ministry say its two counts mean?
9,715 records whose data was not yet complete, against 28,185 records with complete data as of 30 June 2024, is what the Gaza Ministry of Health states in its own post of 24 July 2024. It defines the pair itself: the general martyr count is everyone who reached the hospitals, and the complete-data count is everyone whose data the ministry has finished.
That definition matters because the pair is routinely read as an identified-against-unidentified split. On the ministry’s own wording it is not. It is an administrative state, the difference between a record the Gaza Ministry of Health considers finished and one it does not, and a record can be unfinished for reasons it does not enumerate.
The Gaza Ministry of Health publishes no count of any single missing field: no figure for records missing an identity number specifically, none for a missing date of birth, none for a missing address. The complete-data and not-yet-complete counts are the only completeness figures the issuing body puts out, so any claim about one particular missing field is not the ministry’s claim.
The Gaza Ministry of Health also adds to its cumulative total by administrative approval rather than by new deaths, recording 233 additions on 23 March 2025 and 280 on 30 August 2025 as records approved by a judicial committee. Those are not deaths that occurred on those dates.
All of this is a party to the war reporting on its own record, self-reported and not independently verified, and cited here for what the Gaza Ministry of Health states about itself.
What does the ministry say it does not count?
Five exclusions, stated on 7 January 2024 and never withdrawn. The Gaza Ministry of Health states that it does not count the missing and the bodies trapped under rubble, the unidentified who reached hospitals, the unidentified whose bodies Israel handed over, those buried by their families without passing through a hospital, or the victims in Gaza and North Gaza after its information system stopped in November 2023.
Read together, the five say what kind of object the list is: neither a census of the dead nor a roster of the identified, but a record of deaths that reached a recording system and were resolved within it. The Gaza Ministry of Health names the categories that never reach it.
The fifth exclusion is geographic as well as temporal: Gaza and North Gaza fall out of the ministry’s own registration once its information system stopped in November 2023, so coverage is uneven across the Strip in a way the totals do not show.
The list documents carry their own cut-offs. The closing footnote of the list to 29 March 2024 records that for martyrs in the ministry’s own records, Gaza runs only to 10 November 2023 while the south runs to 29 March 2024.
This is the issuing body’s own scope statement, published by a party to the war and not independently verified. Its usefulness is that it is unusually explicit: a reader setting the ministry’s total beside any other count knows, from the Gaza Ministry of Health itself, named categories the total was never built to hold.
Where does each record on the list say it came from?
5,817 of the 28,185 records on the list to 30 June 2024 carry the Gaza Ministry of Health’s own source label of notification by the martyrs’ relatives, against 22,368 labelled ministry records. The two add to 28,185 exactly. The provenance column is the ministry’s own field, not an outside reconstruction of it.
The earlier round is labelled the same way. The list to 29 March 2024, posted on 3 April 2024, records 21,323 entries and splits into 18,845 ministry records against 2,478 family notifications, and those two add to 21,323 exactly.
Computed here from those published rows, the family-notification stratum moves from 11.6% of the March list to 20.6% of the June list. The share of the record reaching the ministry through relatives rather than through its own registration therefore roughly doubles across three months, on the ministry’s own labelling.
Why that matters is a question about data quality rather than about totals. A hospital record and a family notification are produced by different people under different conditions, and the two strata differ visibly in how they are written, as the next section shows. A count in which one stratum is growing is a count whose internal composition is changing even when its definition does not.
This column is the most useful field on the published lists for anyone weighing the count, because it is the issuing body stating, record by record, which of its two channels produced the entry. It is also the field the next section shows is not always present.
Do the lists carry the same fields from round to round?
The Gaza Ministry of Health’s list to 30 April 2024 carries six columns where its lists to 29 March, 30 June and 31 August 2024 each carry seven. The April round drops both the date-of-birth column and the source column and adds an address column, and the identity column is headed one way in March and a different way in June and August.
The seven-column schema runs serial number, name, identity or document number, date of birth, sex, age, and source. The six-column April schema runs serial number, name, document number, age, address, and sex. For that round there is no provenance split to read, because the field carrying it is absent.
The heading of the identity column also changes. In the March list it is the word for identity; in the June and August lists it is the phrase for document number. A reader building a series across rounds is matching a field whose label the ministry itself has changed.
Date formats differ by source stratum inside a single document. In the March list, rows labelled ministry records print full four-digit years while rows labelled family notifications print two-digit years with dashes. In the August list the oldest decedents carry two-digit years a mechanical reader takes for the current decade, which the age column in the same row contradicts.
None of this establishes that any record is right or wrong. It describes what the published documents contain, and it bears on what can be computed from them without the Gaza Ministry of Health’s own arithmetic.
Does any published list carry a date of death?
None of the four list documents read on 10 September 2026 carries one. All four published by the Gaza Ministry of Health hold a serial number, a name, an identity or document number, a date of birth, sex, age and a source, and the Arabic phrase for date of martyrdom appears in no column header of any of them.
The consequence is structural. A name-by-name list with no date-of-death field cannot be sliced into a daily series, matched to a dated event, or compared period by period against any other dated record. The cumulative totals in the ministry’s daily statistical report are a separate document with a separate construction.
The Gaza Ministry of Health has stated why the field is weak. Its clarification of 29 July 2024 records that the date of death was registered either on the family’s report or on the date the body reached the hospital, and that a committee formed to audit the actual date found martyrs who had been under 18 at death while being registered months later.
That is the issuing body describing a known defect in its own record, published by a party to the war and not independently verified. It also explains why the two channels produce different data quality: a family notification carries the date the family reported, a hospital record the date the body arrived.
The practical rule is narrow. Any figure presented as the ministry’s count of deaths in a particular month or around a dated event cannot have come from the published lists, because the field it would require is not on them.
Do the child and women counts share the total’s denominator?
13,319 child martyrs, stated “with complete data till October 7, 2024” in the Gaza Ministry of Health’s own Health Annual Report, Palestine 2024, is printed beside a total of 43,587 with complete data at 11 December 2024 and 45,553 at 31 December 2024. The numerator’s base sits two to three months before either total.
The same section publishes three further category counts on the same October base: 7,216 female martyrs, 3,447 elderly and 1,057 healthcare workers. Any share computed by dividing one of them into a total printed beside it is therefore a share whose base is months out of alignment with its numerator.
The categories also partition nothing. The first three sum to 23,982, no men’s category is published alongside them, and the healthcare-worker count overlaps all three, since a healthcare worker may also be counted as female or as elderly. A bar chart is the only honest shape for them, because a part-to-whole chart would present four overlapping counts as slices of one pie.
The later annual report changes the form again. For 2023 to 2025 the Gaza Ministry of Health publishes shares rather than counts, at 30% children, 16% women, 7% elderly and 47% men, against a stated total of 71,271. A report reading the two editions together is reading two different constructions of the same categories.
This is the mechanism behind any movement in a published women-and-children share, and it is a mechanism rather than a correction. The question of what each share actually counts is the subject of its own report.
Does the ministry’s arithmetic reconcile with its own daily report?
A 23-record gap, on the ministry’s own two numbers. The one place the Gaza Ministry of Health does the subtraction itself, its post of 24 July 2024, states a complete-data count and a not-yet-complete count whose sum lands on the total its daily report published the following morning rather than on the date it names.
The three documents, as the ministry published them:
| Ministry document | Date it names | Cumulative total it carries |
|---|---|---|
| Daily statistical report | 30 June 2024 | 37,877 |
| Daily statistical report | 1 July 2024 | 37,900 |
| Post of 24 July 2024 | 30 June 2024 | 28,185 complete, 9,715 not yet complete |
Source: Gaza Ministry of Health daily statistical reports and its post of 24 July 2024, each the ministry’s own count, published by a party to the war and not independently verified.
Adding the ministry’s own pair gives 37,900, its total for 1 July. A reader subtracting 28,185 from the 30 June total of 37,877 instead gets 9,692, not the 9,715 the Gaza Ministry of Health states. The addition is the ministry’s; the subtraction is this report’s.
The same slip runs through the annual report, whose period-end pair of 45,553 killed and 108,379 injured for “between October 7, 2023, and December 31, 2024” are the figures the ministry published on 1 January 2025, against its own 45,541 for 30 December.
Neither observation makes a figure wrong. Both mean a ministry period-end figure should be read as the next morning’s daily report until it is checked.
Is the one peer-reviewed test independent of the ministry’s list?
22,368 and 5,817, the same two numbers on both documents. The two ministry input lists that Jamaluddine and colleagues describe in their capture-recapture study, hospital morgue records and a rolling online mortality survey launched on 1 January 2024, are precisely the two strata the ministry’s own list to 30 June 2024 decomposes into under its source column, for the identical window.
The Lancet published that study in 2025, peer-reviewed, and Jamaluddine and colleagues estimate 64,260 deaths due to traumatic injury between 7 October 2023 and 30 June 2024, with a 95% confidence interval of 55,298 to 78,525. It is the only peer-reviewed estimate ever produced against this registry.
Its third list is manually scraped social media obituaries. Two of the three inputs are therefore the ministry’s own channels, which means the estimate measures the completeness of that record rather than counting the dead independently of it. Jamaluddine and colleagues present it as a completeness test (Jamaluddine and colleagues, 2025).
It is also narrower than its headline suggests: it covers traumatic injury alone, excluding deaths from healthcare disruption, food insecurity and disease, and its confidence interval spans more than 23,000 deaths.
The lock between the two halves of this report sits here. The historical claim rests on an uncited sentence about wars no study has examined, and the one study that has examined this war was built from the two strata the Gaza Ministry of Health prints in its own provenance column. On this record the count has never been measured against a record system separate from itself.
Methodology
This report reads named documents and describes what each measured and contains.
Four kinds of source are drawn on. A correspondence letter in a peer-reviewed journal, editorially reviewed rather than refereed (Huynh, Chin and Spiegel, 2024). A peer-reviewed research article (Jamaluddine and colleagues, 2025). A peer-reviewed chart review (Vaktskjold and colleagues, 2016). And the official records of the Gaza Ministry of Health, a party to the war reporting its own count: every ministry figure here is self-reported and none independently verified.
An international news explainer is cited for what it recorded, not as a measurement. The counterpart tolls inside its number pairs were produced by a United Nations humanitarian office and are named only as what the Associated Press explainer of 26 October 2023 records.
Four arithmetic operations are the Institute’s own on published rows, marked where they occur: the percentages computed from the number pairs, the same three on the alternative denominator, the two family-notification shares, and the subtraction of the complete-data count from the daily total.
Three limitations bound it. The ministry’s list documents are published only through its media office channel and not on any ministry web host, so they were read from a public mirror whose record counts and column sets match the ministry’s own description. The literature search behind the absence finding was run on 10 September 2026 across the biomedical and preprint index and the open web, and an absence established by search is bounded by it. Nothing in the sources read establishes whether the pre-2023 counts were accurate against another body’s count, because no study has run that comparison.
Conclusion
So has the Gaza Ministry of Health’s historical accuracy ever been tested? On the record read here the sentence carrying the claim has never been a test. It is one line in a one-page letter, printed without a reference marker, in a piece The Lancet classes as Correspondence, whose authors measured five weeks of 2023 and nothing before.
Every element of the claim is checkable and none of it is a measurement. The quoted endpoints reproduce from three number pairs in a news explainer that computes no percentage, and only on one denominator, the ministry’s own. The 2014 value sits unprinted inside the range, and the literature holds no study that ever ran the comparison. What the Gaza Ministry of Health publishes about itself is concrete by comparison: a complete-data count it defines in its own words, five exclusions it has never withdrawn, a provenance column splitting one list into ministry records and family notifications, a schema that changes between rounds, no date of death on any published list, category counts whose base sits months before the total beside them, and an addition of its own two figures landing on the next morning’s total.
A record that names its own gaps is doing something a slogan cannot do in either direction, and it is the only thing here a reader can check line by line. The one peer-reviewed instrument ever turned on it was assembled from two of its own strata.
Which leaves a question about vocabulary rather than about Gaza. When a count has been called accurate for two decades on the strength of a sentence nobody sourced, what exactly has the word been reporting?
Frequently Asked Questions
What would a test of the ministry’s reliability actually have to compare?
Two counts of the same deaths, over the same window, from record systems that do not feed one another. No such study exists for any pre-2023 conflict, and the one peer-reviewed analysis of the current war drew two of its three lists from the ministry itself (Jamaluddine and colleagues, 2025).
Are the figures Hamas’s own?
The lists and the daily statistical report are published by the Gaza Ministry of Health, a ministry of the authority governing the Strip and a party to the war, and are commonly called Hamas figures. Alongside them the Gaza Ministry of Health publishes a definition of its two counts, five named exclusions, and a provenance column separating its own records from family notifications.
How many records on the ministry’s lists are missing an identity number?
No such figure exists. The Gaza Ministry of Health publishes no count of any single missing field, only the pair it defines itself: 28,185 records with complete data and 9,715 not yet complete as of 30 June 2024.
Can the published lists be used to build a timeline of deaths?
No, because none of the four carries a date-of-death field. Any monthly or event-linked figure attributed to the ministry comes from its daily statistical report, a separate document of cumulative totals rather than named records.
What does the 8.0% figure in the letter refer to?
It is a separate comparison for the 2014 war, between the Gaza health ministry’s count of 2,310 and the 2,125 the Israeli Foreign Ministry reported, that ministry’s own count and not independently verified. Like the others it reproduces only on the Gaza ministry’s denominator (Huynh, Chin and Spiegel, 2024).
Does the capture-recapture estimate settle the question?
It measures something narrower than the question asks. Jamaluddine and colleagues estimate 64,260 deaths from traumatic injury to 30 June 2024, with a confidence interval of 55,298 to 78,525, excluding disease, food insecurity and healthcare disruption. Two of its three lists are the ministry’s own.
Sources
- Associated Press (Debre), 2023. What is Gaza’s Ministry of Health and how does it calculate the war’s death toll? Associated Press, 26 October 2023, updated 6 November 2023. apnews.com. International press, cited for what it printed.
- Huynh, Chin, Spiegel, 2024. No evidence of inflated mortality reporting from the Gaza Ministry of Health. The Lancet, 403(10421), pp. 23 to 24. DOI 10.1016/S0140-6736(23)02713-7. PMID 38070526. Correspondence in a peer-reviewed journal, editorially reviewed and not peer-reviewed.
- Jamaluddine and colleagues, 2025. Traumatic injury mortality in the Gaza Strip from Oct 7, 2023, to June 30, 2024: a capture-recapture analysis. The Lancet, 405(10477), pp. 469 to 477. DOI 10.1016/S0140-6736(24)02678-3. PMID 39799952. Peer-reviewed.
- Ministry of Health, Gaza, 2024 to 2026, ongoing. The published martyr lists of 3 April, 5 May, 24 July and September 2024, the daily statistical report, and the completeness statements of 7 January, 24 July and 29 July 2024. Published through the ministry media office channel. Official record published by a party to the war, self-reported and not independently verified.
- Ministry of Health, Palestine, 2025. Health Annual Report, Palestine 2024, section 1.1, and Health Annual Report, Palestine 2025. site.moh.ps. Official record published by a party to the war, self-reported and not independently verified.
- Vaktskjold and colleagues, 2016. Mortality in the Gaza Strip. Conflict and Health. DOI 10.1186/s13031-016-0077-6. Peer-reviewed.