What’s At Stake In the Privatization of Public Health?

A new National Academy of Medicine Discussion Paper, "Public Health COVID-19 Impact Assessment: Lessons Learned and Compelling Needs" paints a grim picture of the fragmented landscape of public health infrastructure in America: 

"Overall funding for foundational capabilities has run dry in the face of long-standing neglect and deprioritization by both local and national leaders, with the expenditures of public health agencies decreasing by approximately 10% (between 2010 and 2018) and the share of health care spending attributable to public health declining by nearly 17% (between 2002 and 2014). Indeed, rather than valuing prevention, the American system has become increasingly biased in favor of reaction, with per capita spending on public health services equivalent to 1-3% of per capita expenditures on medical care. Chronically deprived of resources, the capabilities of health departments have begun to atrophy over several key domains." 

YIKES. But upon a few moments of further reflection, one realizes the alarm bells about the state of public health infrastructure have been ringing for over 30 years. Reports since the 1980s have addressed the lack of uniform data standards and systems across the states, duplicative reporting, fragmented public health surveillance efforts, and the limitations of existing public health funding models. See, for example: IOM 1988 Milbank 2001 IOM 2002.  

What's at stake in the current moment? What we’ve seen in the pandemic is a mismatch between an expansive conception of what governmental public health should provide and a narrow remit to deliver based on shrinking budgets, an aging workforce, and outdated technology. And yes, federalism, which has shaped variations in the practice and funding landscape across the states and territories in consequential but hard to measure ways. Misaligned incentives and competing priorities are hard to overcome. 

Into the gap pivots technology companies, in what Tamar Sharon has termed, "sphere transgressions". The risk? "The crowding out of essential spherical expertise, new dependencies on corporate actors for the delivery of essential, public goods, the shaping of (global) public policy by non-representative, private actors and ultimately, the accumulation of decision-making power across multiple spheres." 

The impact of privatization on the public health sector warrants more consideration than it has received. Two studies published in 2002 examined The Perceived Impact of Privatization on Local Health Departments and Services Privatized in Local Health Departments as characterized by local health department directors. Overall impressions of impact were equivocal: 50% of respondents felt privatization helped in the performance of core public health functions, while 38% felt either privatization hindered or were unsure. Those who felt privatization hindered performance cited concern about the loss of control over performance of privatized functions, and concern that the agency's authorities would be undermined. 

Despite the relative paucity of research on the impact of privatization of public health, I believe there is reason to revisit the issue as a matter of public trust. During the COVID-19 pandemic tech companies such as Apple and Google have demonstrated interest in collaborating with public health agencies, multimillion dollar (and pound) contracts were inked between governments and private sector giving access to health data without opportunity for public deliberation, and the tech sector shows increasing interest in healthcare more broadly.  

Chiara Cordelli frames the problem this way: "If private actors are morphing into government, can they act with the legitimacy that government claims? And can a government morphed into a network of private actors still govern those subject to its rules legitimately?" From the public's perspective, perhaps one of the greatest threats in the privatization of public goods is loss of a seat at the table and the ability to hold the government to account.