Doubt is not a pleasant condition, but certainty is an absurd one.
—Voltaire (in a letter to Frederick The Great, 1767)
OUR RELATIONSHIP IS BASED on mutual trust. Physician and patient share a common aim: the best medical care, transacted in confidence, that circumstances permit. Thus, we implicitly believe our patients and vice versa. With sexually transmissible diseases (STD), this tacit contract can easily be compromised. The physician could feel uncomfortable asking the tough questions; the patient could fear physician opprobrium and disease-reporting requirements. Both individual and public health outcomes can be damaged by inaccurate perceptions on either side of the therapeutic relationship. No shortcoming, however, is likely to impact the validity of scientific medicine as profoundly as misreporting of information by patients or its uncritical acceptance by the physician. Here, we remind the physician that although trust is good, trust and verification is better. Moreover, it is essential for the validity of evidence-based medicine. Two examples serve to illustrate our concern with low evidentiary standards.
First, in a recent report published in a reputable medical journal, the authors conclude that the detection of anal human papillomavirus (HPV) in male intravenous drug users (IDU) who deny a history of homosexual activity argues for a new mode of HPV transmission: autoinoculation. 1 The authors dismiss the likelihood of their HIV-infected IDU patients lying about receptive anal intercourse, simply because these patients had not previously revealed this behavior to their physicians! In brief, these physicians trust but fail to verify. Two decades of the HIV epidemic has generated sufficient evidence that lying to one’s physician is not uncommon, 2,3 particularly by members of populations who have a reputation for manipulative and dishonest behavior and a documented high prevalence of antisocial personality disorder. 4 Despite caveats in the literature, the authors fail to use appropriate methodology and fail to seek the evidence necessary to assess their patients’ self-reports. 3,4 They did not enroll non-HIV-infected, non-IDU heterosexual control subjects; as common as HPV infection is, seeking evidence of autoinoculation from a genital to an anal locus from such comparisons is methodologically crucial. They fail to record clinical data suggestive of receptive anal intercourse such as examination for lax sphincter tone and for the presence of fissures, fistulae, lesions, or trauma. 5,6 The questionnaire they use apparently does not contain psychometric validity checks 3,4 and was not administered by means of computer-assisted self-interview (ACASI), a mechanism known to substantially enhance reporting of behaviors viewed as pejorative. In one study, for example, reports of anal intercourse increased by 800% with the use of ACASI technology. 7 Finally, the authors fail to collect data on imprisonment, not only a common experience for IDU, but also a milieu where anal rape occurs frequently. Indeed, a recent review of U.S. studies conservatively estimates that at least 10% of detainees are raped. 8
Second, attribution of HIV acquisition to “heterosexual transmission” (a rubric that most people assume to mean vaginal intercourse, despite the notable prevalence of men having anal intercourse with women 4,6) has long been the default category used by epidemiologists in developed countries when HIV-infected patients deny histories of anal (especially receptive) intercourse, IDU, or iatrogenic exposure. Yet, the very low probability of HIV transmission from vaginal intercourse in otherwise healthy persons has been demonstrated in many ways, 4 perhaps most compellingly by an experimental study that showed that when inundated with HIV, intestinal tissue was easily infected, whereas unpunctured vaginal or cervical tissue was not. 9 Notably, attempts to support this default category by examining HIV-infected “heterosexuals” for markers of high-risk behaviors (eg, anal pathology or/and hepatitis C testing) are seldom made (“Patients may lie, but biomarkers don’t” idea) for fear of offending the patient or because of budget considerations. Under such constraints, the cost to epidemiologic validity is out of proportion to its price. Because such validity checks are likely to underestimate the behavior for which they are indicators, multiple checks are necessary to evaluate the transmission route with confidence.
It could be that patients prefer disclosing to misreporting; however, perceptions and conditions that usually accompany physician–patient transactions in the evaluation of sexually transmitted disease/HIV infection serve to discourage candor. Hence, a high index of suspicion for nonconventional behaviors, a willingness to critically assess patients’ self-reports, and relevant detective work, are critical elements for buttressing scientific validity, if not trust. Absence of validity checks in scientific reports should convince the reader to remain agnostic on the issues, and in the absence of such validity checks, the claims of autoinoculation of HPV and HIV from vaginal transmission in healthy persons share the same evidentiary status as gonorrhea acquired from toilet seats. 10,11 Tools are available to boost our confidence in evidence-based medicine and epidemiology. They need to be used. In clinical medicine, when trials of novel therapies are conducted, the likelihood of misreport is often assessed by monitoring blood levels of the candidate medication, by counting the number of pills returned, or by other methods. 12,13 In STD/HIV epidemiology, similar standards need to be routinely implemented. Epidemiology by default assumption or by assertion belongs to a scientifically primitive age.
1. Piketty C, Darragh TM, Costa MD, et al. High prevalence of anal human papillomavirus infection and anal cancer precursors among HIV-infected persons in the absence of anal intercourse. Ann Intern Med 2003; 138: 453–459.
2. Potterat JJ, Phillips L, Muth JB. Lying to military physicians about risk factors for HIV infections. JAMA 1987; 257: 1727.
3. Brody S. Patients misrepresenting their risk factors for AIDS. Int J STD AIDS 1995; 6: 392–398.
4. Brody S. Sex at Risk. Lifetime Number of Partners, Frequency of Intercourse, and the Low AIDS Risk of Vaginal Intercourse. New Brunswick, NJ: Transaction Publishers, 1997.
5. Sohn N, Robilotti JG. The gay bowel syndrome. Am J Gastroenterol 1977; 67: 478–484.
6. Brody S, Potterat JJ. Assessing the role of anal intercourse in the epidemiology of AIDS in Africa. Int J STD AIDS 2003; 14: 431–436.
7. Gross M, Holte SE, Marmor M, Mwatha A, Koblin B, Mayer KH. Anal sex among HIV-seronegative women at high risk of HIV exposure. The HIVNET Vaccine Preparedness Study 2 Protocol Team. J Acquir Immun Defic Syndr 2000; 24: 393–398.
8. Human Rights Watch. No Escape: Male Rape in US Prisons. New York: Human Rights Watch, 2001.
9. Greenhead P, Hayes P, Watts PS, Laing KG, Griffin GE, Shattock RJ. Parameters of human immunodeficiency virus infection of human cervical tissue and inhibition by vaginal virucides. J Virol 2000; 74: 5577–5586.
10. Gilbaugh JH, Fuchs PC. The gonococcus and the toilet seat. N Engl J Med 1979; 301: 91–93.
11. Rein MF. Nonsexual acquisition of genital gonococcal infection. N Engl J Med 1979; 301: 1347.
12. Farmer KC. Methods for measuring and monitoring medication regimen adherence in clinical trials and clinical practice. Clin Ther 1999; 21: 1074–1090.
13. Brody S, Preut R, Schommer K, Schurmeyer TH. A randomized controlled trial of high dose ascorbic acid for reduction of blood pressure, cortisol, and subjective responses to psychological stress. Psychopharmacology 2002; 159: 319–324.