National Newspaper Portrayal of U.S. Nursing Homes: Periodic Treatment of Topic and Tone

Abstract

Context

Although observers have long highlighted the relationship of public distrust, government regulation, and media depictions of nursing-home scandals, no study has systematically analyzed the way in which nursing homes have been portrayed in the national media. This study examines how nursing homes were depicted in four leading national newspapers—the New York Times, Washington Post, Chicago Tribune, and Los Angeles Times—from 1999 to 2008.

Methods

We used keyword searches of the LexisNexis database to identify 1,704 articles pertaining to nursing homes. We then analyzed the content of each article and assessed its tone, themes, prominence, and central actor. We used basic frequencies and descriptive statistics to examine the articles’ content, both cross-sectionally and over time.

Findings

Approximately one-third of the articles were published in 1999/2000, and a comparatively high percentage (12.4%) appeared in 2005. Most were news stories (89.8%), and about one-quarter were on the front page of the newspaper or section. Most focused on government (42.3%) or industry (39.2%) interests, with very few on residents/family (13.3%) and community (5.3%) concerns. Most were negative (45.1%) or neutral (37.0%) in tone, and very few were positive (9.6%) or mixed (8.3%). Common themes were quality (57.0%), financing (33.4%), and negligence/fraud (28.1%). Both tone and themes varied across newspapers and years.

Conclusions

Overall, our findings highlight the longitudinal variation in the four widely read newspapers’ framing of nursing-home coverage, regarding not only tone but also shifts in media attention from one aspect of this complex policy area to another. The predominantly negative media reports contribute to the poor public opinion of nursing homes and, in turn, of the people who live and work in them. These reports also place nursing homes at a competitive disadvantage and may pose challenges to health delivery reform, including care integration across settings.

Despite progress improving the quality of nursing homes over the last two decades, significant challenges remain (Miller and Mor 2006, 2008; Wiener, Freiman, and Brown 2007). Just one-quarter of Americans rank nursing-home operators as “very high” or “high” in honesty and ethics, while only one-third believe that nursing homes are doing a “good job,” both figures falling far below those for other health care providers (Jones 2010; Kaiser Family Foundation 2001, 2005, 2007). More than half of long-term care opinion leaders ranked the quality of care provided by the average nursing home as “fair or poor,” compared with one-quarter to one-third of the leaders who ranked assisted-living facilities, hospitals, and home care agencies this way, just 14.2 percent for adult day care providers, and 5.8 percent for hospice workers (Miller, Clark, and Mor 2010).

The mass media can play a significant role in shaping people's views of the importance and nature of particular issues (Baumgartner and Jones 1993; Dearing and Rogers 1996; Iyengar and Reeves 1997; Kingdon 1995; McCombs, Shaw, and Weaver 1997). Media scholars refer to both first-order agenda-setting effects and second-order agenda-setting effects. First-order effects are the media's influence on people's ranking of the relative importance of general topics or issues, and second-order effects pertain to the manner in which specific topics or issues are framed.

Framing involves emphasizing some aspects of an issue or topic rather than others. It is important because varying the emphases can elicit very different reactions in the reader, regarding not only the issue's salience but also the attributes of the issue themselves (Entman 1993). Ghanem (1997) characterizes framing along four dimensions: the prominence with which a story is presented (e.g., placement, story format), the affect of the story (e.g., positive, negative, neutral), the specific subtopics addressed, and other attributes (e.g., the central actor). Characterizing how the media reports on particular issues is important because the choices made can result in varying assessments of the seriousness of a problem, its antecedents and solutions, and the motivation of key actors.

Hundreds of media studies have been reported in the agenda-setting literature (Dearing and Rodgers 1996; McCombs and Reynolds 2002). The large majority have demonstrated first-order agenda-setting effects, most frequently in relation to the public agenda (Barabas and Jerit 2009; Iyengar, Peters, and Kinder 1982; McCombs and Shaw 1972), but also in relation to public officials and the government agenda (Baumgartner and Jones 1993; Kingdon 1995; Wanta and Kalyango 2007), and sometimes both (Gonzebach 1992; Leff, Protess, and Brooks 1986; Tan and Weaver 2007). Others delve more deeply, demonstrating second-order effects, again with respect to both the public agenda (Gollust, Lantz, and Ubel 2009; Iyengar and Simon 1993) and the government agenda (Baumgartner and Jones 1993; Soroka 2002).

Eighty percent of Americans report hearing or reading about nursing homes (Kaiser Family Foundation 2001), and 60 percent are exposed to at least one news story during the course of a year (Kaiser Family Foundation 2001). A sizable proportion, 21 to 27 percent, report basing their impression of nursing homes mainly on what they have seen or heard in the media (Kaiser Family Foundation 2001, 2005). The relationship of public distrust, government regulation, and media depictions of nursing homes has been discussed anecdotally for decades (Hawes 1987; Smith 1981; Vladeck 1980). This is reflected in Vladeck's (1980, 203) observation more than thirty years ago that “nursing home issues only get onto the political agenda when propelled there by scandals or other untoward events” reported in the media.

Despite the potential linkage of media portrayals, public opinion, and government action, few scholars have sought to understand the way in which nursing homes have, in fact, been portrayed, none in relation to tone or over the last ten years. Smith (1981) traced reports on nursing homes in the New York Times from 1956 through 1978, concluding that the marked increase identified between 1974 and 1977 dealt primarily with criminal-related activities. Ulsperger (2002) traced the transformation in media representation of the National Citizens Coalition for Nursing Home Reform in six U.S. newspapers, with reporting framed largely around resident-specific issues (e.g., general abuse, restraint use) (1987–1992), followed by a lull in coverage (1993–1994) and framing largely around structural concerns (e.g., billing, staffing) (1995–1999). Mebane (2001) examined the content of long-term care coverage during 1998 by twenty-two U.S. newspapers. Nursing homes dominated, with 38 percent of long-term care articles devoted to this topic. Of these, 38 percent reported on business matters, 23 percent on quality, and 19 percent on justice system issues. Just 13 percent appeared on the front page, and few underlying issues—population aging, costs, regulation—were discussed.

Given the absence of recent research investigating the media's coverage of the nursing-home sector, we decided to look at how nursing homes were depicted in four widely read national newspapers—the New York Times (NYT), Washington Post (WP), Chicago Tribune (CT), and Los Angeles Times (LAT)—from 1999 to 2008. Through a content analysis of 1,704 articles on nursing homes, we documented the kinds of coverage nursing homes receive in terms of tone and content and how this has changed over time. Is the media coverage of the nursing-home sector predominately negative, as is often anecdotally claimed? Has the volume and nature of that coverage varied over time in light of key events impacting the nursing-home field? Has it varied across the newspapers examined?

This article begins with a review of major policy events and trends in nursing-home quality, financing, litigation, and disaster responsiveness over the last fifteen years. This is followed by a description of study methods and results, both overall and across newspapers, by tone, and over time. It concludes with a discussion and our conclusions, addressing the major implications of this study for both long-term care and health policy and politics more generally.

Major Policy Events and Trends in the Nursing-Home Sector

Nursing homes are the most identifiable setting for long-term care. Currently, there are approximately 1.7 million beds in about 16,000 nursing homes in the United States (Harrington, Hauser, et al. 2011). Between 1999 and 2008, expenditures on nursing homes increased from $90.7 billion to $138.4 billion (CMS 2005, 2010).The primary purchaser is Medicaid, which constituted 40.6 percent of total nursing-home spending in 2008, followed by out-of-pocket payments (26.7%), Medicare (18.6%), and private insurance (7.4%). Figure 1 is a time line highlighting major events in the nursing-home sector from 1995 to 2010.

FIGURE 1.

Major events in nursing-home sector, 1995–2010.

Quality

New, higher standards of care and revisions of the federal nursing-home enforcement process promulgated with the Omnibus Budget Reconciliation Act (OBRA) of 1987 went into effect in 1995 (Miller and Mor 2008; Wiener, Freiman, and Brown 2007). Despite these reforms, the late 1990s and early 2000s witnessed a proliferation of studies and reports identifying problems and strategies for improving the quality of nursing-home care (Health Care Financing Administration 2000; Institute of Medicine 2001; Office of the Inspector General 1999; U.S. General Accounting Office 1999). It was also during this period, 1998, that a resident assessment instrument, also mandated by OBRA 1987, was implemented nationwide. This has, in turn, informed public report cards, including the Nursing Home Compare website, which the Centers for Medicare and Medicaid Services (CMS) launched in 2002 to give the public information about the quality of care provided (Castle and Ferguson 2010). Beginning in 2002, the CMS's Quality Improvement Organizations also were directed to consult with nursing homes to develop and implement quality improvement projects (Quality Partners of Rhode Island 2005).

During this period, the “culture change” movement began to take hold as well (Rahman and Schnelle 2008). Culture change requires care providers to adopt the architectural redesigns, organizational processes, and environmental improvements necessary to achieve person-centered care (Koren 2010). Although the movement began slowly, it then gained momentum, in part owing to the September 2006 launch of the Advancing Excellence in America's Nursing Home campaign, a voluntary collaboration of nursing homes, government officials, and other stakeholders to support participating facilities’ quality improvement goals (Advancing Excellence 2010). In 2008, the CMS added a five-star rating system to Nursing Home Compare so consumers would have an easier way to distinguish between high- and low-performing facilities. The 2010 Patient Protection and Affordable Care Act (ACA) also included several provisions meant to improve transparency in the nursing-home sector.

Financing

The Balanced Budget Act (BBA) of 1997 repealed the federal Boren amendment, which had reduced state government discretion regarding Medicaid's nursing-home reimbursement, based on litigation under the amendment's provisions (Miller 2008, 2011). The BBA also transformed nursing-home reimbursement under Medicare, from a reasonable, cost-based, retrospective system to one based on a prospectively determined payment rate, adjusted for patient acuity and other factors (i.e., case mix). Not only did this change lead to a sudden, dramatic decline in Medicare outlays, but it also reduced the profitability of ancillary services, resulting in financial difficulties for many operators (Harrington, Carrillo, et al. 2011). Partly because of these perceived industry hardships, the federal government enacted Medicare “give-backs” with the Balanced Budget Refinement Act of 1999 and the Benefits Improvement and Patient Protection Act of 2000, which temporarily restored some of the per diem reimbursement that had been lost.

In 2005, the Deficit Reduction Act expanded the Long-Term Care Partnership Program, which allows individuals in participating states to protect substantially more assets when qualifying for Medicaid, if they had previously purchased a private policy, in addition to tightening Medicaid's eligibility rules related to asset transfers and home equity protection. Then the onset of the Great Recession in 2008 resulted in across-the-board retrenchment at the state level. Thus, despite $87 billion in additional federal Medicaid funding provided under the economic stimulus, the number of states reducing or freezing nursing-home payments increased markedly (Smith, Gifford, and Ellis 2010). There also was a sharp decline in retirees’ and workers’ confidence that they would have enough money to pay for long-term care services (Helman et al. 2010). The ACA included the Community Living Assistance Services and Supports (CLASS) Act, a federally administered, voluntary, long-term care insurance program first introduced in Congress in 2005 (Miller 2010). But the implementation of CLASS was halted owing to concerns about the program's solvency (Wiener 2012).

Litigation

When compared with the rate of litigation against other providers, that against nursing homes for negligence has historically been quite low. But this began to change in the mid- to late 1990s, particularly in states like Texas and Florida, which, for example, accounted for more than half of such litigation in 2001 (Stevenson and Studdert 2003). Although the rate of litigation has since declined somewhat, 4,716 claims were filed against 1,465 facilities in the nation's five largest nursing-home chains between 1998 and 2006, 61 percent of which resulted in payouts averaging $199,794 per claim, most commonly for falls, pressure ulcers, dehydration/malnutrition, physical/verbal abuse, and medication errors (Studdert et al. 2011). In general, poorer-performing nursing homes are at a greater risk of being sued, though the evidence suggests that the claimants exercise limited discrimination in this regard (Johnson et al. 2004; Studdert et al. 2011).

Natural Disasters

Especially salient during the period of our study were Hurricanes Katrina and Rita (Barnes et al. 2008). In September 2005, thirty-five elderly men and women drowned at St. Rita's Nursing Home in St. Bernard's Parish in New Orleans during Hurricane Katrina. Two years later, the owners were charged but were later acquitted of criminally negligent homicide and cruelty for not evacuating their residents. But St. Rita's was not the only nursing home to fail to evacuate because of logistical challenges and the frailty of their patients. Twenty-two elderly residents of Lafon Nursing Home of the Holy Family died in the days following the hurricane, largely because of the slow and disorganized emergency response to the storm. Soon thereafter, the risk associated with evacuating was highlighted when twenty-three frail residents died when the bus they were riding caught fire and exploded as they fled Hurricane Rita. Together, the two hurricanes highlighted the potential drawbacks of both evacuating and staying in place (Dosa et al. 2007).

Methods

Data

Trained research assistants retrieved all the articles on nursing homes published in the four study newspapers from January 1, 1999, to December 31, 2008, from the complete commercial version of the LexisNexis database. They used the following search terms: nursing home / nursing homes, long-term care facility / long-term care facilities, and nursing facility / nursing facilities. Each article was screened for relevance. Those excluded from the final sample were duplicates (including news briefs), obituaries, advertisements, and articles that were not about nursing homes because, say, they mentioned nursing homes only in passing or were about another topic (e.g., reporting on a local election site). Of approximately 5,000 articles identified, 1,704 were kept for analysis.

Coding Strategy

We developed a coding instrument to systematically abstract information from each of the articles identified. Our initial set of coding categories was refined through an inductive process in which two of the investigators read nursing home–related articles and independently suggested categories that they observed in the data. These categories were then further refined over several iterations, in collaboration with other members of the research team (Glaser and Strauss 1967; Miles and Huberman 1984). This process culminated in the final coding instrument.

One of the study investigators trained the research assistants (RAs) to use the coding instrument using a sample of nursing home–related articles different from those analyzed in this study. That is, they were supposed to come up with codes consistent with those of the other RAs and the study investigator. In instances in which there was disagreement, they discussed the articles in the context of the coding instrument until consensus was achieved. Intercoder reliability tests were conducted five times. Each interrater reliability check had a sample of fifty articles on nursing homes. With experience, the overall level of agreement reached 85 percent (range = 75% to 100% across individual categories). Once the four RAs had been properly trained, they began coding the four newspapers. To ensure rigor, we conducted biweekly meetings to discuss their progress with the coding and their application of coding categories, as well as to resolve any difficulties with coding specific articles.

Content Abstracted

In addition to determining the overall volume of coverage, both cross-sectionally and over time, we examined its tone, prominence, themes/subthemes, and central actor.

Affect

The information about affect included the following mutually exclusive categories: positive, negative, neutral, or mixed. Articles with a positive tone portrayed the nursing-home sector in a favorable light, leaving readers “feeling good” by the end. They also relied on strongly positive adjectives and anecdotes. Examples are “The New Nursing Home, Emphasis on Home” and “Singing Just for Seniors; A Volunteer Troupe in L.A. Takes Its Talents to Nursing Homes.” Articles with a negative tone, by contrast, emphasized less desirable aspects, leaving readers feeling negatively by the end. They also employed strongly negative adjectives and anecdotes. Examples are “At Nursing Home, Katrina Dealt Only with First Blow; Nuns Labored for Days in Fatal Heat to Get Help for Patients” and “13 Nursing Homes Accused of Abuse, Fraud in Suit.” Articles with a neutral tone were factual pieces that eschewed strong wording, personal statements, or anecdotes. These did not elicit strong emotion from the readers because of how the story was presented and worded. Examples are “Elder Care Company Sold for $1.9 Billion” and “Planning for Nursing Home Care.” Articles with a mixed tone presented both positive and negative aspects of a story, reporting a range of reactions and/or pitting two sides against each other. These stories often included both positive and negative wording in approximately equal measure. Examples are “Hope and Humor Got New Orleans Nursing Home Residents through the Storm, but Four Died along the Way” and “Some Dying Patients Have Neither Friends nor Family. Increasingly, Volunteers Are Filling In.”

Prominence

The information on prominence included the article type (news, editorial, column, letter) and location (front page, front section, elsewhere). The coding categories for both article type and location were mutually exclusive.

Themes/Subthemes

The information on themes was on the following general topics: quality, financing, negligence/fraud, cost, business/property, home- and community-based services, natural disasters, and access. Because the themes were not mutually exclusive, the articles could be coded under more than one theme, as appropriate. Subthemes also were identified for quality, financing, and negligence/fraud, which also were not mutually exclusive.

Quality refers to the quality of care experienced by residents and efforts to address or improve it, including government reports about the state of quality assurance in the nursing-home sector. Examples are “Nursing Homes Cited, Fined in Patient Deaths” and “Nursing Safety Net Frayed, Report Says; Staffing Shortages, Poor Conditions at Facilities Are Called the Greatest Threat to Patients.” When appropriate, these articles were also coded according to the following subthemes: person-centered care/culture change (articles about making nursing homes more homelike/inviting to residents, e.g., the Advancing Excellence Campaign, Quality Improvement Organization involvement); consumer reporting (articles about making information on nursing homes more accessible to residents/families, e.g., Nursing Home Compare, the five-star rating program, the ACA's oversight provisions); staffing (articles about the level and quality of staffing and efforts to improve it, e.g., staff recruitment and retention, mandatory staffing ratios); regulation (articles about maintaining or improving the quality and performance of nursing homes, e.g., the implementation and modification of OBRA 1987's quality-of-care provisions); and safety (articles about health or safety issues in nursing homes, e.g., fire, food poisoning).

Financing concerns the sources of payment for nursing-home care, including Medicare, Medicaid, private insurance, family/self-funded care, and the CLASS program. Examples are “Long-Term Care Insurance: How Much Is Too Much?” and “Panel Calls for Big Changes in Medicaid.” When appropriate, these articles were also coded according to the following subthemes dealing specifically with Medicaid (e.g., reductions in nursing-home reimbursement, impact of federal fiscal relief, asset transfers, home equity protection), Medicare (e.g., implementation of prospective payment under the BBA), private insurance (e.g., the Long-Term Care Partnership Program, barriers to take-up), and family/self-funded care (e.g., a reduction in savings rates due to the Great Recession).

Negligence/fraud refers to legal proceedings concerning maltreatment, fraud, misconduct, and abuse, including prevailing trends in nursing-home litigation during the time period studied. Examples are “Medicaid Fraud Ill Received” and “Nursing Homes Face Prosecution.” When appropriate, these articles were also coded according to the following subthemes: malpractice/negligence (articles about facility malpractice or resident neglect); elder abuse (articles about physical, sexual, or emotional abuse); fraud (articles about defrauding Medicare, Medicaid, and other payers); and liability claims, torts, and lawsuits (articles about legal actions taken against nursing homes).

Cost is how much it costs to be a client at a nursing home or to run a facility. Examples are “The Nursing Home with Everything (for a Price)” and “Getting Stuck with the Tab, Tighter Asset Spend-Down Rules Will Force More Families to Cover Nursing Home Costs Alone.”

Business/property means any of a number of business and real estate issues, including the downsizing, expansion, repair, maintenance, purchase, or opening of a facility. Examples are “Health Care Property Investors Downgraded” and “Nursing Home Chain Sells for $4.9 Million.”

Home- and community-based services (HCBS) refers to diverting or moving residents from nursing homes to home- and community-based settings. Examples are “Home Care Services for the Elderly” and “Alternatives to the Nursing Home.”

Natural disasters pertain to the impact and response (or lack thereof) of operators, government, and other parties to natural disasters, including Hurricanes Katrina and Rita and their aftermaths. Examples are “Bus Evacuating Senior Center Burns, Killing 24 near Dallas” and “Report Calls for Tighter Rules in Nursing Home Evacuations.”

Access is the processes that might prevent or enhance access to nursing-home care, including waiting lists, bed shortages, provider capacity, and other pertinent issues. Examples are “Clinton OKs Limits on Nursing Home Evictions” and “Facilities Struggle to Treat Alzheimer's Patients; Few Nursing Homes Are Able to Give Proper Psychiatric Care: A New Campaign Hopes to Change That.”

Central Actor

The central actor is the dominant protagonist, the individual or organizational entity whose behavior is the primary focus of the actions described in the following mutually exclusive categories: government, industry, residents/families, and community. Central actor was coded as “government” if the primary action involved local, state, or federal politicians; departments; agencies; or judicial bodies (e.g., “In Effort to Pare Medicaid Rolls, Long-Term Care Is the Focus”). It was coded as “industry” if the primary action involved a specific nursing home, nursing-home staff, or nursing-home industry more generally (e.g., “Nursing Homes Face Prosecution”). It was coded as “residents/family” if the primary action pertained to one or more current or prospective nursing-home residents and/or their families (e.g., “My Mother and Her Nursing Home”). Finally, it was coded as “community” if the primary action concerned members of the broader community, such as volunteer agencies, religious entities, local nonprofits, and schools (e.g., “Students See the Future, and It's Elderly People”).

Data Analysis

We used basic frequencies and descriptive statistics to describe the content of the articles, followed by cross-tabulations, Pearson χ²-tests, Oneway ANOVAs, and graphing to describe the relationships among the variables analyzed, focusing especially on newspaper, year, and tone. We present our overall findings first, followed by trends across tone, by newspaper, and over time. Only statistically significant findings are mentioned in the text; χ²-tests and p-values are reported for those findings not also reported in table 1.

TABLE 1.

Proportion of Nursing-Home Articles, Totals, and Article Characteristics by Tone (n= 1,704)

	Total % (n)	Positive % (n)	Neutral % (n)	Negative % (n)	Mixed % (n)	χ2-statistic (d.f.)	p-value
Newspaper
Los Angeles Times	43.4 (740)	10.5 (78)	35.0 (259)	42.2 (312)	12.3 (91)	48.1 (9)	< .001
Washington Post	6.7 (114)	8.8 (10)	36.8 (42)	54.4 (62)	0.0 (0)
Chicago Tribune	16.3 (227)	12.3 (34)	33.9 (94)	50.5 (140)	3.2 (9)
New York Times	33.6 (573)	7.3 (42)	41.0 (235)	44.3 (254)	7.3 (42)
Year
1999	19.0 (324)	13.6 (44)	37.3 (121)	39.8 (129)	9.3 (30)	76.4 (27)	< .001
2000	16.0 (272)	9.2 (25)	47.4 (129)	37.5 (102)	5.9 (16)
2001	8.5 (144)	4.2 (6)	52.1 (75)	40.3 (58)	3.5 (5)
2002	7.0 (120)	15.0 (18)	36.7 (44)	43.3 (52)	5.0 (6)
2003	9.2 (156)	8.3 (13)	26.9 (42)	53.2 (83)	11.5 (18)
2004	7.2 (122)	8.2 (10)	28.7 (35)	50.0 (61)	13.1 (16)
2005	12.4 (212)	6.1 (13)	35.4 (75)	49.1 (104)	9.4 (20)
2006	7.6 (129)	7.0 (9)	34.1 (44)	48.8 (63)	10.1 (13)
2007	7.5 (127)	11.8 (15)	27.6 (35)	49.6 (63)	11.0 (14)
2008	5.8 (98)	11.2 (11)	30.6 (30)	54.1 (53)	4.1 (4)
Article Type
News Story	89.9 (1,531)	9.0 (138)	39.5 (605)	43.6 (668)	7.8 (120)	58.8 (9)	< .001
Editorial	6.7 (114)	12.3 (14)	11.4 (13)	61.4 (70)	14.9 (17)
Letter	2.2 (37)	18.9 (7)	8.1 (3)	59.5 (22)	13.5 (5)
Column	1.3 (22)	22.7 (5)	40.9 (9)	36.4 (8)	0.0 (0)
Location
Front Page	11.2 (191)	8.4 (16)	23.6 (45)	53.4 (102)	14.7 (28)	30.7 (6)	< .001
Front Section	13.4 (228)	10.5 (24)	31.6 (72)	50.4 (115)	7.5 (17)
Elsewhere	75.4 (1,285)	9.6 (124)	39.9 (513)	42.9 (551)	7.5 (97)
Number of Themes
1	56.2 (958)	12.6 (121)	39.2 (376)	37.4 (358)	10.8 (103)	68.7 (6)	< .001
2	37.0 (640)	5.7 (36)	34.8 (219)	54.4 (343)	5.1 (32)
3+	6.8 (116)	6.0 (7)	30.2 (35)	57.8 (67)	8.3 (142)
Themes
Quality	57.0 (972)	23.4 (108)	34.7 (160)	29.7 (137)	12.1 (56)	102.2 (3)	< .001
Financing	33.4 (569)	2.3 (5)	55.4 (118)	30.0 (64)	12.2 (26)	66.0 (3)	< .001
Negligence/Fraud	28.1 (478)	0.7 (1)	18.9 (28)	75.7 (112)	4.7 (7)	112.9 (3)	< .001
Cost	10.0 (170)	0.0 (0)	44.8 (13)	48.3 (14)	6.9 (2)	5.4 (3)	.146
Business/Property	6.3 (107)	2.0 (1)	82.4 (42)	5.9 (3)	9.8 (5)	44.7 (3)	< .001
HCBS	5.8 (89)	33.3 (6)	27.8 (5)	22.2 (4)	16.7 (3)	8.6 (3)	.036
Natural Disasters	5.3 (90)	0.0 (0)	4.8 (1)	81.0 (17)	14.3 (3)	20.4 (3)	< .001
Access	3.1 (52)	0.0 (0)	36.4 (4)	54.5 (6)	9.1 (1)	2.3 (3)	.506
Central Actor
Government	42.3 (720)	4.9 (35)	45.7 (329)	39.7 (286)	9.7 (70)	206.7 (9)	< .001
Nursing Home	39.2 (668)	9.0 (60)	33.2 (222)	52.5 (351)	5.2 (35)
Residents/Families	13.3 (226)	12.4 (28)	27.0 (61)	46.9 (106)	13.7 (31)
Community	5.3 (90)	45.6 (41)	20.0 (18)	27.8 (25)	6.7 (6)
	100 (1,704)	9.6 (164)	37.0 (630)	45.1 (768)	8.3 (142)

Note: Analyses of theme by tone are limited to the 958 single-themed articles.

Findings

Overall

Volume

Most of the 1,704 articles analyzed were published in the LAT (43.4%) and the NYT (33.6%), with substantially fewer in the CT (16.3%) and the WP (6.7%) (table 1). More than one-third of the articles (35.0%) were published in 1999 and 2000, and a comparatively high percentage (12.4%) in 2005 as well (table 1, figure 2).

FIGURE 2.

Annual volume of nursing-home coverage by newspaper, selected themes, and tone.

Tone

The tone of most nursing-home coverage was typically negative (45.1%) or neutral (37.0%); it was very rarely positive (9.6%) or mixed (8.3%) (table 1).

Prominence

Substantially more nursing-home articles were news stories (89.9%) than editorials, columns, or letters (10.2%). Approximately one-quarter appeared either on the front page of the newspaper (11.2%) or a section inside the paper (13.4%).

Themes

The most frequently identified general themes were quality (57.0%), financing (33.4%), and negligence/fraud (28.1%) (table 1). Of the quality-related subthemes identified, regulation appeared the most frequently (20.2%), followed by safety (15.4%) and staffing (12.7%), then person-centered care / culture change (6.3%) and consumer reporting (3.2%). Of the financing-related subthemes used, Medicaid (14.6%) and Medicare (10.2%) appeared the most frequently. In contrast, there was very little discussion of private insurance (2.0%) and self-financing (0.5%). Finally, of the negligence/fraud or legal subthemes employed, malpractice/negligence (17.0%) appeared the most frequently, followed by legal claims, torts, and lawsuits (12.0%); elder abuse (7.3%); and fraud (4.2%). Most of the articles (56.2%) were coded according to just one of the primary themes used; however, a little more than a third (37.0%) were coded according to two themes, and fewer than one in ten (6.8%), according to three or more.

Central Actor

The central actors identified were typically government officials or institutions (42.3%) or the nursing-home industry, in general or with respect to a particular facility (39.2%) (table 1). Less frequently, the central actors were nursing-home resident(s)/family (13.3%) or the broader community (5.3%).

By Newspaper

Tone

Nursing home coverage was more negative in the WP (54.4%) and the CT (50.5%) than in the LAT (42.2%) and the NYT (44.3%), the latter two in which, at 12.3 percent and 41.0 percent, respectively, it was more often mixed and neutral than in the other newspapers (table 1).

Prominence

Opinion pieces (primarily editorials) were twice as common in the NYT (16.1%) as in the LAT (9.9%) and the WP (7.0%); none were identified in the CT (χ²= 54.5, p < .001). The CT was most likely to place an article on the front page (20.9%); the WP (21.9%) and the LAT (20.0%), at the front of a section (χ²= 108.1, p < .001).

Themes

The WP (62.3%) and the CT (66.8%) were more likely to report on quality than were the NYT (53.8%) and the LAT (44.1%) (χ²= 15.6, p < .001). This is in contrast to the NYT (48.2%) and the LAT (31.4%), which were substantially more likely to report on financing than were the other newspapers (< 16.0%) (χ²= 113.4, p < .001), and the CT (47.3%) and the WP (34.2%), which were substantially more likely to report on negligence/fraud than their counterparts (< 25.0%) (χ²= 67.3, p < .001).

Central Actor

There was some variation in the central actor across the newspapers. The WP (6.1%) was less likely to address resident/family issues than the other newspapers were (12.6% to 15.9%) but was more likely to address community issues (14.0% vs. 0.9% to 7.2%) (χ²= 53.9, p < .001).

By Tone

Prominence

Editorials and letters, at 61.4 percent and 59.5 percent, respectively, were much more likely to be negative than were news stories and columns (< 45.0%), which were much more likely to be neutral (> 39.0% vs. < 12.0%) (table 1). By contrast, columns (22.7%) and letters (18.9%) were more likely to be positive than editorials and news stories (< 13.0%). In general, articles posted on the front page of the newspaper (53.4%) or section (50.4%) were especially likely to be negative, followed by those placed elsewhere (42.9%).

Themes

Articles characterized by one theme were less likely to be characterized negatively (37.4%) and more likely positively (12.6%) than articles characterized by two (54.4%/ 5.7%) or three or more (57.8%/ 6.0%). Articles on business/property (5.9%) and HCBS (22.0%) were least likely to be characterized negatively (table 1). This is in contrast to articles about negligence/fraud (75.7%) and natural disasters (81.0%), which were most likely to be presented in a negative light. Relative to other themes, articles about quality were comparatively likely to be described both positively (23.4%) and negatively (29.7%). At 33.3 percent, articles about HCBS were particularly likely to be characterized positively, and at 82.4 percent, articles about business/property issues in a neutral manner.

Central Actor

Community articles were much more likely to be positive than articles focusing on other central actors (45.6% vs. < 13.0%) (table 1). By contrast, industry articles (52.5%) were more likely to be negative than resident/family (46.9%), government (39.7%), and community (27.8%) reports. Government articles (45.7%) were more likely to be neutral than industry articles (33.2%), which were more likely to be so than resident/family (27.0%) and community (20.0%) pieces.

Over Time

Tone

General tone varied over time, with the proportion of negative portrayals, for example, rising from 39.8 to 53.2 percent between 1999 and 2003, after which it remained steady at approximately 50.0 percent, before peaking at 54.1 percent in 2008 (table 1, figure 2). By comparison, neutral coverage was more predominant during the early years, characterizing a majority of articles (52.1%) in 2001. In all but two years (2000 and 2001), however, the modal tone was negative.

Prominence

Articles were especially likely to be on the front page of the newspaper in 2005 (21.2%) and the front page of a section in 2001 (22.9%) and 2008 (24.0%) (χ²= 62.3, p < .001). At 22.2 percent in 1999, and 12.9 percent and 11.8 percent in 2000 and 2007, respectively, the proportion of opinion pieces exceeded the proportion in other years (< 8%) (χ²= 83.9, p < .001).

Themes

The proportion reporting on quality rose steadily until 2003 (67.3%) where it peaked before declining through 2006 (48.1%), after which it increased through 2008 (63.3%) (figure 2). The proportion of reporting on financing, by contrast, peaked in 1999/2000 (41.0%/ 42.6%) and 2008 (41.8%), while the proportion of reporting on negligence/fraud increased through 2004 (from 21.3% to 37.7%), after which it remained steady (30.7% to 37.2%) before declining in 2008 (26.5%). The proportion of reporting on natural disasters peaked in 2005 (23.1%) before decreasing every year afterward.

The overall tone of articles characterized by the three most prevalent themes varied over time (figure 3). The proportion of quality articles with a negative tone peaked in 2003, at 55.2 percent; thereafter, it remained above 50.0 percent before declining somewhat in 2008 (43.5%). There also were periods when the proportion of positive articles about nursing-home quality increased as well, including 1999/2000 (18.9%/ 16.8%), 2002 (23.0%), and 2007/2008 (19.1%/ 16.1%). While the proportion of financing articles with a positive tone remained small throughout the entire study period (< 10.0%), the proportion with a negative tone oscillated over time. Thus, after decreasing from 37.5 to 31.7 percent between 1999 and 2001, the proportion with a negative tone reached 52.5 percent in 2003 before bottoming out at 25.0 percent in 2005 and peaking at 68.3 percent in 2008. In contrast, the proportion of negligence/fraud articles with a positive or a negative tone did not vary significantly.

FIGURE 3.

Proportion of nursing-home articles per year by theme and tone of coverage.

Central Actor

The proportion of articles focusing on particular central actors varied over time (figure 4). Between 1999 and 2002, more articles focused on government than on industry actors, an emphasis that flipped between 2003 and 2005 before reversing once again during the remaining years (χ²= 43.0, p < .026). By contrast, the proportion of community and resident/family articles remained relatively small throughout, although there was an uptick in resident/family reports in 2008, concomitant with a decline in government and industry reports.

FIGURE 4.

Proportion of nursing-home articles per year by central actor.

Discussion

This article analyzed the content of 1,704 newspaper articles to assess how nursing homes were depicted in four widely read newspapers—the New York Times, Washington Post, Chicago Tribune, and Los Angeles Times—from 1999 to 2008. Our results support widespread anecdotal claims that the tone of media coverage has been primarily negative, rarely positive, and otherwise neutral. Although certain issues—quality, financing, and negligence/fraud—were commonly discussed, others, such as natural disasters, spiked dramatically at critical moments. Both the tone and extent of coverage about particular themes varied over time and across newspapers.

The media coverage of nursing homes tends to be much less extensive than the media coverage of other issues (Mebane 2001; Pew Research Center 2004–2012). This is the case for health policy/systems issues more broadly, just a small proportion of which is devoted to long-term care. From January 2007 to June 2009, for example, health care constituted approximately 6 percent of total newspaper coverage, with about 40.0 percent devoted to health policy/systems issues, and the remainder to public health and specific diseases/conditions (Kaiser Family Foundation and Pew Research Center 2008, 2009). Although the health reform debate became a dominant issue, at 9.8 percent, in the year or so preceding enactment of the Patient Protection and Affordable Care Act (ACA) (June 2009–March 2010), the proportion of newspaper coverage devoted to the issue dropped to 1.9 percent after the ACA was signed into law (April 2010–March 2011) (Pew Research Center 2010, 2011). Among the more dominant non-health-related issues covered during the time period studied were the economic crisis, campaigns/elections, terrorism, and the Iraq and Afghanistan wars; among the more dominant health-related issues were Medicare prescription drugs, health care costs, health reform, and the uninsured.

The relatively small proportion of total newspaper coverage devoted to long-term care is consistent with the general finding that Americans do not rank the issue among the nation's most pressing problems or priorities in health care or otherwise (Kaiser Family Foundation 2001–2012; Pew Research Center 1998–2012). For example, in 2000, when asked what the top priority for health reform should be, only 8 percent answered “lack of long-term care options,” far behind HMOs, costs, and Medicare prescription drugs (16% to 26%) (Hart and Teeter Research Companies 2000). And in 2009, when asked what the top priority for health reform should be, just 3 percent said “providing stronger, affordable long-term care options,” far below coverage, cost, fraud/waste, and insurance (13% to 19%) (AARP 2009).

Nonetheless, a large proportion of Americans believe that long-term care is an important problem. This is reflected in a May 1998 survey in which 88 percent of the respondents stated that long-term care for the elderly was an important issue facing the nation (Kaiser Family Foundation 1999); in a May 2001 survey in which 63 percent believed providing more funding for nursing homes was an important issue (Harvard 2001); and in a December 2007 survey in which 86 percent reported being concerned about the nursing-home care available (Kaiser Family Foundation 2007). Together these surveys suggest that although the general public does not rank long-term care among the nation's top priorities, it is nonetheless an issue to which they pay some attention at least some of the time and about which they hold generally negative views (Jones 2010; Kaiser Family Foundation 2001, 2005, 2007). The amount and nature of media coverage may play a key role in this regard.

As noted, media scholars refer to both first-order agenda-setting effects (what people think about) and second-order agenda-setting effects (how people think). The overall volume of coverage was particularly high in some years (1999/2000 and 2005), suggesting that the saliency of nursing homes as a general issue may have varied over time with respect to both the general public and government officials. This finding is consistent with prior research demonstrating marked changes in the volume of coverage related to nursing homes (during the 1970s) and other areas (Baumgartner and Jones 1993; Pew Research Center 2010, 2011; Smith 1981; Taylor-Clark et al. 2007; Times Mirror 1995).

The spike in 2005 clearly reflected the dramatic rise in reporting on natural disasters stemming from Hurricanes Katrina and Rita. The subsequent drop in disaster-related coverage indicates that the storms and their immediate aftermaths were deemed more newsworthy than the subsequent investigations and changes made to address deficiencies in federal, state, and local emergency operations systems and institutional policies and procedures (Brown, Hyer, and Polivka-West 2007). Less clear is why there was substantially more coverage evident during 1999/2000 than in subsequent years. One possibility is that our data reflect the more general decline in newspaper circulation and coverage during this time period owing to the rise in digital, cable, and other alternative news media (Pew Research Center 2004–2012). Another is that the data reflect the comparatively large number of newsworthy events during those years, particularly in the areas of nursing-home financing and quality. The former includes the aftermath of the BBA and its adverse impact on the nursing-home industry. The latter includes the aforementioned government studies and reports and associated congressional hearings identifying continuing quality problems in nursing homes despite efforts to improve government oversight and regulation in this area (Health Care Financing Administration 2000; Institute of Medicine 2001; Office of the Inspector General 1999; U.S. General Accounting Office 1999).

Baumgartner and Jones (1993, 103) demonstrated “how the media play an integral role in the policy process by directing attention alternately toward different aspects of the same issue over time.” The basic notion is that public perceptions of an issue can change dramatically owing to the media's tendency to frame the same issue in different ways at different times. This appears to be the case in our data, although the focus of the balance of coverage in most years was negative depictions of quality. First, we observed that the proportion of quality articles increased through 2003 before declining through 2006, after which it increased once again through 2008. Interwoven between these peaks in quality reporting were disproportionately high levels of reporting on financing (1999–2002), negligence/fraud (2003–2007), and natural disasters (2005–2006). Thus, the relative proportion of reporting on quality varied somewhat with the rise and fall in reporting on other aspects of the issue over time. There was also variation in the relative emphasis placed on the central actors identified, with government predominating early on (1999–2002), the industry during the middle years (2003–2005), and government once more toward the end (2006–2008), when a rise in reporting on resident/family issues was evident as well. Trends such as these are important because, as Baumgartner and Jones (1993) observed, shifts from one topic or emphasis to another can markedly influence the general tone of coverage and, in turn, overall perceptions.

Our findings suggest that nursing-home coverage was less negative during the early years of our time series (1999–2002), due to relatively high levels of neutral reporting about financing combined with somewhat more positive/neutral (though still largely negative) reporting about quality. Indeed, a high proportion of reporting during this time period reflected neutral depictions of actions by the federal government designed to support and document nursing-home financing and quality. This includes the Medicare “give-backs” enacted in light of perceived hardships faced by the nursing-home industry. It also includes the series of government reports on nursing-home quality, the launch of Nursing Home Compare, and the inaugural Quality Improvement Organization consultation with nursing homes.

Our findings suggest that nursing-home coverage became more negative during the middle to late years of the time series (2003–2007), partly because of the comparatively high levels of negative reporting about negligence/fraud and the Gulf Coast hurricanes, a larger proportion of which focused on industry than on government-related activities and performance. That such a large proportion of coverage was devoted to negligence/fraud, particularly during the middle years of the time series, supports prior research and anecdotal claims regarding the perceived newsworthiness of nursing-home scandals (Mastin et al. 2007; Smith 1981; Vladeck 1980). Hurricane coverage pertaining specifically to nursing homes was overwhelmingly negative, whereas other articles in the aftermath of the storm varied in tone, based on who was assigned responsibility for the disasters (Barnes et al. 2008). Clearly, nursing-home operators were blamed for the deaths of elderly nursing-home residents because of their lack of preparation.

The last year of the time series, 2008, is particularly interesting. Although reporting on negligence/fraud and, especially, natural disasters, declined to comparatively low levels, the proportion of negative articles hit its peak at 54.1 percent, largely because of a marked increase in the proportion of negative articles about financing, from just 25.0 percent in 2005 to 68.3 percent in 2008. This suggests maintenance of the predominately negative tone of coverage through a shift in the topics addressed; that is, from natural disasters to events surrounding the Great Recession, including a lack of adequate preparation for the costs of long-term care and reductions and freezes in Medicaid nursing-home payments. Furthermore, while negative reporting about quality continued to predominate in 2008, the proportion of neutral articles about this topic increased, thereby supplementing the relatively high proportion of positive articles reported since 2007. Several high-profile events account for these trends, including implementation of the CMS's five-star rating system and the launch of the Advancing Excellence in America's Nursing Home campaign by the nursing-home industry and its collaborators.

In addition to longitudinal variation, potentially important differences in coverage across the newspapers regarding volume, themes, prominence, and tone were identified. Differences such as these could stem from variation in the frequency of newsworthy events (e.g., bankruptcies, lawsuits, budgetary cutbacks). They could also stem from editorial assessments of audience receptivity and interest relative to other issues. Differences in the norms, resources, and priorities of the individual newspapers we examined could play a role as well. Frequently highlighted in the literature are the variations in advocacy efforts targeted at advancing public policy through the strategic use of the mass media (Stillman et al. 2001; Wallack and Dorfman 1996). Nursing-home resident advocates in some states may be more effective at cultivating relationships with journalists and providing them with credible and timely information and feedback than are advocates in other states. That this dynamic may help explain the variation in reporting observed in our data is suggested by prior research indicating that resident advocacy groups in some states have been particularly influential in raising awareness of nursing-home issues and promoting policy change advantageous from the consumer point of view, in part through their media advocacy efforts (Miller 2008, 2011). Indeed, among the most noteworthy consumer advocacies are the Long Term Care Community Coalition in New York and the California Advocates for Nursing Home Reform in California. Both these organizations are affiliates of the Washington, DC–based National Consumer Voice for Quality Long-Term Care (formerly the National Citizens Coalition for Nursing Home Reform), which has played a key role in championing federal nursing-home regulation, in part by highlighting continuing deficiencies in nursing-home quality nationwide (Edelman 1997/98; Ulsperger 2002).

Limitations

To our knowledge, ours is the most comprehensive study so far to examine the way in which nursing homes have been portrayed in the national media. There are several limitations worth noting, however. First, our findings may not be generalizable to other newspapers or media. The New York Times, the Washington Post, and the Los Angeles Times were chosen because, with circulations of 876,638, 600,449, and 545,345, respectively, they are among the top five most widely circulated dailies in the United States (Audit Bureau of Circulations 2010). The Chicago Tribune was chosen to ensure more balanced regional representation; with a circulation of 441,508, it is the most widely circulated newspaper outside California and New York. Research suggests that the proportion and nature of coverage may vary across newspapers with different circulations (large, medium, small), markets (national, regional, local), audiences (general, niche), and ownership (Barnes et al. 2008; Payne, Appel, and Kim-Appel 2008; Pew Research Center 1998–2012). It also indicates that the amount and nature of coverage may vary across news outlets, for example, newspaper, print magazine, network news, PBS, talk radio, and cable news (Kaiser Family Foundation and Pew Research Center 2008, 2009; Mebane 2001; Pew Research Center 2010; Times Mirror 1995).

Second, a handful of general search terms were used to identify the articles analyzed. Since we did not exhaust all possible terms, we may have missed some potentially relevant articles.

Third, the coders may have interpreted the information abstracted differently, despite efforts to promote consistency through uniform training, regular meetings, and periodic reliability checks. This is more likely with more subjective items such as tone than with more objective items such as article type and location.

Fourth, we used conservative decision-making rules when coding tone. In order for an article to be coded as negative, it had to highlight undesirable attributes or events and employ strongly negative adjectives and anecdotes. Thus, articles that highlighted potentially negative events (e.g., high costs, bankruptcies) but that were written factually were typically coded as neutral. Accordingly, we may have overestimated the number of articles with neutral content and underestimated the number of articles with negative content.

Fifth, although we identified and abstracted a comprehensive array of article characteristics, other potentially useful dimensions were not abstracted but may be worth examining, for example, author/source, use of quotations, inclusion of photographs/images, and episodic versus thematic framing.

Sixth, this study relied on a content analysis of a large number of nursing-home articles. In-depth qualitative analysis might have permitted a more nuanced examination of the media coverage reported by, say, applying qualitative coding techniques to a subsample of articles addressing particular topics of interest. Using these methods, future research should aim to understand what cultural stereotypes and values underlie the ways in which newspaper articles speak about and portray nursing home–related issues. It should also examine the extent to which the media give readers a greater understanding of the underlying problems facing the nursing-home sector, as well as follow-ups on stories exposing the sector's problems to determine what impact, if any, they had on government, consumer, and provider behavior.

Seventh, we examined media framing, but not the way in which it impacts government policy. Future research should examine the relationship between media coverage and changes in public policy related to nursing homes, both across jurisdictions—federal, state, and local—and over time, controlling for other socioeconomic, programmatic, and political characteristics using appropriate quantitative or statistical techniques.

Conclusions

Overall, our findings highlight varied framing of nursing-home coverage in four widely read national newspapers over time, not only in tone but also in shifts in media attention from one aspect of this complex policy area to another. Moreover, the predominantly negative coverage reported is consistent with ongoing problems in the nursing-home sector. This includes challenges associated with understaffing, negligence, fraud, and deficiencies in quality of care (Miller and Mor 2006, 2008; Wiener, Freiman, and Brown 2007). It also includes increasing frailty, disability, medical complexity, and segregation of nursing-home residents into higher- and lower-quality facilities on the basis of race, ethnicity, location, and payment status (Feng et al. 2006, 2011).

A negative portrayal of the nursing-home sector has implications for both nursing-home residents and providers. It seems reasonable to suppose that media reports about nursing-home negligence and fraud, quality-of-care scandals, and regulatory inadequacies have contributed to ongoing public scrutiny and efforts to improve the way that nursing homes are regulated, including modifications of the survey and certification process and improvements in quality measurement and public reporting. It also seems reasonable to suppose that media reports have contributed to the negative public opinion of nursing homes and, in turn, of the people who live and work in them. The “second rate” status associated with caring for elderly people in nursing homes is one reason that facilities have trouble recruiting and retaining sufficient levels of staff, whether physicians, nurses’ aides, or other workers (Mickus, Luz, and Hogan 2004; Miller and Mor 2007). Stereotyping, bias, and discrimination of old people and other norms are among the reasons why opportunities for civic engagement by nursing-home residents are typically limited (Anderson and Dabelko-Schoeny 2010; McBride 2006/2007). Lack of individual preparation for the high costs of nursing-home care, whether through savings or long-term care insurance purchases, has been partly attributed to psychological barriers associated with avoidance and denial (Helman et al. 2010; Melnyk 2005; Pinquart and Sorensen 2002). The negative coverage of the nursing-home sector has helped reproduce and reinforce the negative stereotypes and norms underlying each of these dynamics.

Consumer behavior has been affected as well. Nursing homes face growing competition from noninstitutional alternatives to institutional placement, including home- and community-based services and assisted-living facilities. Between 1999 and 2007, Medicaid home- and community-based expenditures and participants increased from $17 billion to $42 billion and 1.9 million to 2.8 million people (Ng and Harrington 2011). This growth has been driven, in part, by policymakers’ belief in the cost-effectiveness of these programs (Wiener 2010). It also has been driven in response to consumers’ preferences to remain at home and in the community as long as possible (Kaiser Family Foundation 2007). In 2007, there were an estimated 11,276 assisted-living facilities, with 839,746 units (Stevenson and Grabowski 2010). Like HCBS, growth in assisted living has also been driven by consumers’ preferences. Unlike HCBS, it has served as a substitute for nursing-home care for a primarily privately paying cohort of individuals (Grabowski, Stevenson, and Cornell 2012). The media's substantiation of the poor reputation of the nursing-home industry has further placed such facilities at a competitive disadvantage relative to more highly rated HCBS and assisted-living providers (Jones 2010; Miller, Clark, and Mor 2010).

The contributions of adverse media coverage to the negative reputation of the nursing-home industry could also pose challenges to health delivery reform, including the ultimate success of ACA provisions promoting care integration across settings through the development of Accountable Care Organizations (ACOs) and bundled payments. To be successful, the kind of care that nursing homes provide will need to become an integral part of these initiatives. If postsurgical rehabilitation is to be included in an illness episode, for example, the hospital performing the surgery will need to coordinate with the participating nursing home to achieve the desired savings and outcomes. But while it will be necessary for hospital systems, physicians’ groups, and other health care providers to work closely with nursing homes, the large health care systems on which the ACO concept is partly based do not currently include nursing homes in their provider networks to any meaningful extent (Berenson and Burton 2011; Higgins et al. 2011). This means that to develop ACOs, health care systems will need to establish new relationships with nursing homes, by either contracting with or buying one or more facilities. As noted, hospitals, physicians’ groups, and health plans are viewed much more favorably by the general public than nursing homes are (Jones 2010; Kaiser Family Foundation 2001, 2005, 2007), possibly because of the media's reports of ongoing problems in the nursing-home sector.

How will these disparities in public opinion affect the success of ACOs, bundled payments, and other new arrangements targeted at improving care integration? Might more favorably viewed provider modalities have concerns about linking up with less reputable nursing-home providers, for either their own reputations or performance? How eager will hospitals, physicians’ groups, and other potential partners be to integrate nursing homes into their systems? The answers to questions like these remain to be seen but may depend on how nursing homes are portrayed in the national media. This, of course, will hinge largely on the extent to which the industry and government address the underlying problems that result in the disproportionately negative media attention in the first place.