Deidentification of Patient Data

By | Uncategorized | No Comments

In the United States, the Health Insurance Portability and Accountability Act of 1996 (HIPAA) governs the use and disclosure of protected health information (PHI). Under HIPAA, entities like healthcare providers must carefully protect PHI from disclosure. However, providers are not barred from sharing other kinds of health information. In fact, HIPAA contains guidelines for the release of health information, which facilitates data sharing, commerce, and research. If the identifying information is removed from patient data through a process called deidentification, HIPAA protections no longer apply, and entities are free to use or disclose such deidentified health information.


Drawing on statistics, algorithms, and machine learning, academic researchers have devised many methods for deidentification of patient data. For example, some methods use manually crafted patterns to classify what information is “PHI-like” or “not PHI-like”. Other methods use supervised machine learning to automatically determine patterns as to what qualifies as PHI [1]. Regardless, the end goal is the same: deidentified data must be immune to reverse-engineering (called reidentification) that could compromise patient privacy. Under HIPAA, providers are liable if they fail to take reasonable deidentification measures [2]. On the other hand, HIPAA standards strive to minimize the risk of reidentification without unduly burdening providers.


Providers can deidentify data using what HIPAA calls the expert determination method. Under this method, a person with sufficient experience analyzes the data to determine whether the risk of reidentification is “very small” [3]. The expert does not need a specific education credential or professional certification, but they should have experience with statistics, mathematics, data science, or a similar field. To improve accountability, the expert must maintain documentation of their methods and results. By hiring an expert to make these determinations, healthcare providers can have greater assurance that their patient data is HIPAA-compliant.


Without access to a qualified expert, providers can also deidentify data through the safe harbor method. In this case, the provider must remove eighteen different types of data that could identify a patient [3]. These data include names, street addresses, telephone numbers, email addresses, Social Security numbers, full-face photographs, and others. HIPAA contains specific provisions for handling certain kinds of data, like birthdates and ZIP codes. Additionally, providers must take additional steps to remove information that they know could be identifiable [3]. For example, if one patient has a specific and unique occupation title, the provider should not include occupation data in their dataset. If healthcare providers cannot readily hire an expert, performing the safe harbor method in-house should ensure HIPAA compliance.


In summary, US federal law standardizes the handling of protected health information and deidentified health information. Yet, questions remain about whether HIPAA’s standards for the deidentification of patient data are sufficient. In the statistics literature, reidentification has been performed on many kinds of datasets outside the healthcare space. For example, reidentification was performed on an anonymized dataset of Netflix subscribers by cross-referencing Netflix watch data with public IMDB movie/television ratings [4]. Researchers have not systematically studied whether current reidentification methods are effective on patient data, and further research is needed to assist expert determinations and healthcare policymakers [5]. Future research may inform new standards for improving patient privacy or new paradigms for handling patient data.




[1] S. M. Meystre, et al. Automatic De-identification of Textual Documents in the Electronic Health Record: A Review of Recent Research. BMC Medical Research Methodology 2010; 10: 70. DOI: 10.1186/1471-2288-10-70.  

[2] US Department of Health and Human Services. Special Topics: Research. 2018. URL: 

[3] US Department of Health and Human Services. Guidance Regarding Methods for De-identification of Protected Health Information in Accordance with the Health Insurance Portability and Accountability Act (HIPAA) Privacy Rule. 2015. URL: 

[4] A. Narayanan and V. Shmatikov. How To Break Anonymity of the Netflix Prize Dataset. 2006. ArXiv:cs/0610105.  

[5] K. El Emam, et al. A Systematic Review of Re-Identification Attacks on Health Data. PLOS One 2011. DOI: 10.1371/journal.pone.0028071.  

Anesthesia for Cancer Biopsies

By | Uncategorized | No Comments

Choosing what type of anesthesia to use and where to administer it for a cancer biopsy can be challenging. For one, anesthesia providers must take into account the risk factors associated with each patient. Anesthesia can also interact negatively with patients’ monocytes and macrophages, particularly with those cells’ chemotactic and phagocytic functions [1]. Furthermore, anesthesia can also compromise patients’ antigen recognition mechanisms [1]. Accordingly, the choices made by anesthesia providers when performing cancer biopsies should not be taken lightly. This article will address some anesthesia recommendations for three forms of cancer biopsies: lung/abdominal, breast, and prostate.


Image-guided percutaneous needle biopsies (PNBs) are the predominant method of searching for cancer in the lungs and abdomen [2]. Local anesthetic is recommended for most patients undergoing a PNB [2]. Physicians tend to administer lidocaine (from 10 to 20 mL) 1 to 2% along the needle’s predicted route [2]. This method is highly successful. When performed with local anesthesia, PNBs used for non-small cell lung cancer detection have a 95% diagnostic accuracy [3]. Complications are rare, with the most common issue being pneumothorax in 5-10% of cases [3]. Despite local anesthesia’s high success rates, general anesthesia may be more appropriate for children, along with any other patients who might have difficulty following instructions during the procedure [2].


For breast cancer biopsies, researchers have experimented with various forms of anesthesia in recent years [4]. In 2018, Levins et al. investigated the interaction between different forms of anesthesia and the body’s distribution of μ-opioid receptors (MOR), natural killer cells, and other immune cells–all of which are related to cancer prognoses [4]. They did not find a significant difference between a balanced general anesthetic with opioid analgesia and a propofol-paravertebral anesthetic with continuing analgesia in terms of how each affected immune cell marker expression [4]. However, the former method increased resected tumor MOR expression [4]. Because of the correlation between MOR expression and a tumor’s level of aggression, these results may dissuade anesthesia providers from administering general anesthesia [4]. Given the well-documented success of local anesthetics during breast cancer biopsies, a transition away from general anesthesia appears feasible [5].


Prostate biopsies are usually conducted using transrectal ultrasounds (TRUS) [6]. The two most common forms of anesthesia for TRUS are periprostatic nerve block (PNB) and intrarectal topical anesthesia (ITA) [6]. While PNB was considered, for a long time, to be more reliable and successful than ITA in reducing pain during TRUS biopsies, recent evidence has revealed that ITA can also significantly reduce TRUS-related pain, albeit not as well as PTB [6]. Regardless, given their comparably low rates of adverse events, this knowledge could support administering joint ITA and PNB regimens during transrectal ultrasounds [6]. Alternatives to TRUS have also been researched recently, with a study by Stefanova et al. suggesting that transperineal prostate biopsies under local anesthesia could be feasibly adopted [7]. In that study, transperineal prostate biopsies seemed a safer alternative than TRUS, with patients experiencing no more than mild levels of discomfort [7]. Depending on the method chosen, anesthesia providers have multiple options when treating prostate cancer biopsy patients.


While these are just some of the recent discoveries related to anesthesia in cancer biopsies, they reflect the general idea that techniques are evolving and skewing away from general anesthesia when possible.




[1] M. Alieva, J. van Rheenen, and M. L. D. Broekman, “Potential impact of invasive surgical procedures on primary tumor growth and metastasis,” Clinical & Experimental Metastasis, vol. 35, no. 4, p. 319-331, May 2018. [Online]. Available:


[2] L. Dalag, J. K. Fergus, and S. M. Zangan, “Lung and Abdominal Biopsies in the Age of Precision Medicine,” Seminars in Interventional Radiology, vol. 36, no. 3, p. 255-263, August 2019. [Online]. Available:


[3] P. Lavaud et al., “Focus on Recommendations for the Management of Non-small Cell Lung Cancer,” Cardiovascular Interventional Radiology, vol. 42, no. 9, p. 1230-1239, May 2019. [Online]. Available:


[4] K. J. Levins et al., “The effect of anesthetic technique on µ-opioid receptor expression and immune cell infiltration in breast cancer,” Journal of Anesthesia, vol. 32, no. 6, p. 792-796, September 2018. [Online]. Available:


[5] B. M. Saputra et al., “ESRA19-0646 Ultrasound guided erector spinae plane block as regional anesthesia techniques for breast cancer biopsy, a case series,” Regional Anesthesia & Pain Medicine, vol. 44, supp. 1, p. A264-A265, February 2019. [Online]. Available:


[6] Y. Yang et al., “The Efficiency and Safety of Intrarectal Topical Anesthesia for Transrectal Ultrasound-Guided Prostate Biopsy: A Systematic Review and Meta-Analysis,” Urologic Internationalis, vol. 99, no. 4, p. 373-383, December 2017. [Online]. Available:


[7] V. Stefanova et al., “Transperineal Prostate Biopsies Using Local Anesthesia: Experience with 1,287 Patients. Prostate Cancer Detection Rate, Complications and Patient Tolerability,” Journal of Urology, vol. 201, no. 6, p. 1121-1126, June 2019. [Online]. Available:

Human Genome Editing: Ethical Considerations

By | Uncategorized | No Comments

Rapid advances in genome editing technology following the development of CRISPR have led to almost miraculous progress toward the treatment of some diseases, such as sickle cell disease, beta thalassemia, and congenital blindness. These treatments induce somatic gene edits, meaning they are not passed on to offspring, and as such, there is widespread consensus that they should be administered to patients and researched further. Much more controversial is the use of germline editing, in which edits are made to an early-stage human embryo in vitro that is brought to term such that all the baby’s cells, and those of its future descendants, have the edited gene. The greatest concern surrounding this technology is that, once our understanding of the genetics of complex traits like intelligence and personality advance, prospective parents will be able to select which traits their child will inherit, resulting in the normalization of “designer babies.” Given the technology’s enormous potential toward both good and harm, ethical considerations for the young field of human genome editing are a major area of discussion.

While this reality is currently far beyond our technical capabilities, very real fears were realized in 2018 when Chinese scientist He Jiankui violated current regulations and ethics protocols by mutating a cell receptor for HIV from embryos in order to confer HIV resistance before implanting the embryos to be brought to term.1 In the aftermath of this bombshell announcement, discussions about the ethics of germline editing came to the fore of the scientific community. Many human genome editing researchers called on nations to enact a moratorium on germline editing until “broad societal consensus” on the ethical considerations of germline editing could be reached.2 Others have argued that, while caution must certainly be exercised, trials for germline editing should proceed, as it is unethical not to pursue treatments for genetic conditions that cause profound suffering when the technology exists.3

Much of the conversation on the ethics of germline editing centers on the circumstances in which it should or should not be employed. Many have embraced the argument that germline editing should only be used as a “treatment,” implying the existence of a disease or disability that the genome editing addresses, as opposed to as an “enhancement,” where there is no underlying medical need.4 Allowing germline editing in the latter case, according to many, would lead humanity down the slippery slope to eugenics. However, deciding what constitutes a disease or disability is not straightforward. Many forms of deafness, for example, are congenital, and are potential targets for germline editing. But plenty of deaf people lead long and fulfilling lives and feel they’ve gained from being deaf.5 Even sickle cell disease has advantages: carriers for the disease have increased resistance to malaria.6 Due to the complexity of these cases and others like them, there may not always be a consensus on whether it is ethical to edit the genes we perceive as causing disability out of the human genome.

Another ethical issue posed by human genome editing is that of inequality. When germline editing technologies eventually become mainstream, they will almost certainly be prohibitively expensive, at least at the outset. This may “exacerbate inequality and even permanently encode it into our species.”7

This scenario will not be a possibility for some time, due to our current technical capabilities and the strict regulations enacted by rich nations. All but 5 of 96 developed countries in one analysis prohibit germline editing for reproduction without exception,8 and the World Health Organization recently released a series of recommendations for nations to further ensure that gene editing technologies are not misused.9 Gene editing has the power to change the future of our species, and as our abilities to use it improve rapidly, it is critical that nations, institutions, and individual scientists give ethical considerations of genome editing the highest priority.




1 Cyranoski, D. “What CRISPR-Baby Prison Sentences Mean for Research.” Nature News, Nature Publishing Group, 3 Jan. 2020,

2 Lander, E., et al. “Adopt a Moratorium on Heritable Genome Editing.” Nature News, Nature Publishing Group, 13 Mar. 2019,

3 Ayanoglu, F. B. et al. “Bioethical Issues in Genome Editing by CRISPR-Cas9 Technology.” Turkish Journal of Biology, vol. 44, no. 2, 2020, pp. 110–120., doi:10.3906/biy-1912-52.

Human Genome Editing: Status

By | Uncategorized | No Comments

The field of human genome editing has exploded in the past 10 years, thanks to the revolutionary gene editing technology known as CRISPR. In 2012, Jennifer Doudna and Emmanuelle Charpentier elucidated the mechanism of the CRISPR-Cas9 system, an ancient defense system evolved in bacteria to detect and cut the DNA of invading viruses, and showed that it could be leveraged for targeted genome editing in vitro.[i] Feng Zhang and his team at the Broad Institute subsequently showed that CRISPR-Cas9 could successfully modify target genes in mammalian cells.[ii] These two discoveries led to the implementation of CRISPR in many facets of basic research, the formation of new biotechnology companies dedicated to using CRISPR for a variety of purposes,[iii] and the successful use of CRISPR gene therapies in patients with genetic diseases. In recognition of CRISPR’s extraordinary potential, the Royal Swedish Academy of Sciences – which normally bestows Nobel Prizes several decades after a scientific discovery – awarded the 2020 Nobel Prize in Chemistry to Doudna and Charpentier, a mere eight years after their seminal paper, for discovering the genetic tools that “have taken the life sciences into a new epoch.”[iv]

When used for human genome editing, the CRISPR system only consists of three components: a Cas (CRISPR-associated) enzyme, which is likened to “molecular scissors,” a “guide RNA,” a piece of RNA complementary to the target DNA sequence, and a repair template, the desired sequence of DNA. The guide RNA binds to the desired stretch of DNA, allowed Cas to make a double-stranded break in the DNA. The repair template then becomes incorporated into the cell’s genome as it repairs the break. [v]

This basic approach has already been used to effectively treat several genetic conditions. Over the course of the past two years, patients with sickle cell disease and beta thalassemia, monogenic diseases characterized by defective hemoglobin, had stem cells from their bone marrow removed and edited with CRISPR so that they would begin expressing fetal hemoglobin (which normally is turned off shortly after birth), after which they received infusions of these cells.[vi] More than a year after treatment, the patients presented high levels of blood hemoglobin and a drastic reduction in painful symptoms.[vii] Trials are currently underway to treat genetic forms of blindness in vivo by injecting a harmless virus containing the CRISPR machinery into the back of the eye.[viii] A June 2021 study in which researchers eliminated the production of a defective liver protein that can cause fatal side effects showed that CRISPR can be safe and effective when injected directly into the bloodstream, likely an important step for the future of CRISPR therapeutics.[ix]

Congenital diseases represent a significant portion of health conditions faced globally, and many cause serious pain, disability, morbidity, or early mortality. The development of CRISPR technology has resulted in rapid increases in funding, research, and potential ability to treat such diseases via human genome editing. Results so far are promising, and it is likely that further research will improve the accessibility of this technology.



1 Jinek, M., et al. “A Programmable Dual-RNA-Guided DNA Endonuclease in Adaptive Bacterial Immunity.” Science, vol. 337, no. 6096, 2012, pp. 816–821., doi:10.1126/science.1225829.

2 Ran, F. A., et al. “Genome Engineering Using the Crispr-cas9 System.” Nature Protocols, vol. 8, no. 11, 2013, pp. 2281–2308., doi:10.1038/nprot.2013.143.

2 Shaffer, C. “CRISPR Startups Give Genome Editing Several New Twists.” Genetic Engineering & Biotechnology News, 4 Aug. 2020,

4 “The Nobel Prize in Chemistry 2020.”, 7 Oct. 2020,

5 Pak, E. “CRISPR: A Game-Changing Genetic Engineering Technique.” Science in the News, The Graduate School of Arts and Sciences at Harvard University, 31 July 2014,

6 Stein, R. “1st Patients to Get CRISPR Gene-Editing Treatment Continue to Thrive.” NPR, 15 Dec. 2020,

7 Frangoul, H., et al. “CRISPR-Cas9 Gene Editing for Sickle Cell Disease AND Β-THALASSEMIA.” New England Journal of Medicine, vol. 384, no. 3, 2021, pp. 252–260., doi:10.1056/nejmoa2031054.

8 Stein, R. “In a 1st, Scientists Use Revolutionary Gene-Editing Tool to Edit Inside a Patient.” NPR, NPR, 4 Mar. 2020,

9 Ledford, H. “Landmark CRISPR Trial Shows Promise against Deadly Disease.” Nature News, Nature Publishing Group, 29 June 2021,

COVID-19 Transmission by Children

By | Uncategorized | No Comments

Babies and toddlers may be more likely to transmit the virus that causes COVID-19 to others in their households compared to teenagers, a new study has found (Paul et al., 2021). Since early 2020, the coronavirus pandemic has overwhelmed countries all around the world. In the United States, the number of deaths from coronavirus is nearing 642,000, with an increase in cases, hospitalizations, and deaths in recent weeks (CDC). This study, which was published last month in JAMA Pediatrics, does not resolve the ongoing debate over whether infected children are as contagious as infected adults, nor does it suggest children are drivers of the pandemic. The study does, however, demonstrate that very young children can play a role in the transmission of COVID-19.

Researchers at Public Health Ontario, a Canadian public health agency, analyzed Ontarian health records from June 1 to December 31, 2020, and identified 6,280 households in which a child (0-18 years old) was the “index case” – the first person to develop COVID-19 symptoms or test positive for the virus. Then they looked for “secondary cases,” others in the household who got sick in the two weeks after the first child became ill (Paul et al., 2021). In most cases, the chain of transmission stopped with the infected child, but in 27.3% of households, children transmitted the virus to at least one other household member (Anthes, 2021). Another takeaway was that adolescents were most likely to bring the virus into the home, with 14–17-year-olds making up 38% of index cases (Paul et al., 2021). The study’s key finding, however, was that the odds of household transmission of COVID-19 were roughly 40% higher for infections in children 3 or younger compared to children between 14 and 17 (Paul et al., 2021). Behavioral differences might explain this finding, as babies and toddlers often require close contact and hands-on care. “The 0-to-3-year-old child is held differently, is cuddled,” offered Dr. Paul Offit, professor of pediatrics in the Division of Infectious Diseases at Children’s Hospital of Philadelphia (Salzman et al., 2021). And when young children are sick, for example, they cannot be isolated.

This study updates experts’ understanding of COVID-19 transmission risk. Earlier in the pandemic, some scientists suggested the risk of COVID-19 transmission declined with younger age, though this assumption was likely biased by the fact that lockdowns and social distancing limited  social encounters for young children (Choi, 2021). These new findings suggest the opposite, and Dr. Edith Bracho Sanchez, a primary care pediatrician and assistant professor of pediatrics at Columbia University Irving Medical Center, said that the study “just shows how humble we have to be when it comes to children and this virus. We always knew children could get it, could transmit it, and could get sick with COVID,” she continued (Salzman et al., 2021). “I think we’re learning more and more just how much.”

The study was conducted in 2020, before the delta variant emerged, so further research is necessary to understand transmission risk in the context of the variant and other potential variants. The study also took place before vaccines were available, so all household members were unvaccinated (Paul et al., 2021). Still, its findings reinforce the importance of implementing and maintaining mitigation strategies at schools and childcare facilities, especially as a new school year begins and more children are returning to school in-person. Strategies such as frequent cleaning, good ventilation, distancing, and masking when possible are essential. The study also reaffirms the importance of vaccination for all eligible people over 12, especially those that spend time with children (Salzman et al., 2021).




Anthes E. (2021, August 16). Babies and Toddlers Spread Virus in Homes More Easily Than Teens, Study Finds. New York Times.


Centers for Disease Control (CDC). COVID Data Tracker, Updated Daily. U.S. Department of Health and Human Services.


Choi J. (2021, August 19). Younger children more likely to spread COVID-19 to households than older kids. The Hill.


Paul LA, Daneman N, Schwartz KL, et al. Association of Age and Pediatric Household Transmission of SARS-CoV-2 Infection. JAMA Pediatr. Published online August 16, 2021. doi:10.1001/jamapediatrics.2021.2770

Salzman S., Richter Lauren R. (2021, August 16). Younger children more likely to spread COVID-19, study finds. ABC News.