Common concerns.pdf

Common Concerns Amid Diverse Systems: Health Care Experiences In Five Countries The experiences and views of sicker patients are bellwethers for how well health care systems are working. by Robert J. Blendon, Cathy Schoen, Catherine DesRoches, Robin Osborn, and Kinga Zapert PROLOGUE: Researchers worldwide have exhaustively chronicled the ineffi- ciency of the U.S. health care system. The most expensive and technologically ad- vanced health care system in the world yields health outcomes comparable to those of countries with much lower health spending. Some have pointed out structural misallocations of health care assets that hobble the system?s ability to respond to the changing needs of the U.S. population. Molly Coye (Health Affairs, Nov/Dec 2001), for example, has noted ?highly variable patterns of care, wide- spread failure to implement recognized best practices and standards of care, and the persistent inability of provider systems to achieve substantive changes in pat- terns of practice.? The paper that follows reports findings from the 2002 Commonwealth Fund International Health Policy Survey of Sicker Adults in the United States, Canada, the United Kingdom, Australia, and New Zealand. This study reports the recent experiences of sicker adults in each country with respect to care coordination, physician-patient interaction, medical errors, prescription drug issues, and access to needed care. Interestingly, despite clear structural differences among the sys- tems, findings in all five countries reveal consistent dissatisfaction among sur- veyed populations with general health system quality, stemming from problems associated with medical errors, inadequate patient-physician communication, and insufficient coordination of care. Robert Blendon and colleagues conclude that sizable dividends (improved health and reduced costs) could be garnered from targeting health reform interventions in each country to such subpopulations of sicker adults, particularly those using multiple physicians and medications. Blendon is a professor of health policy and political analysis at the Harvard School of Public Health. Cathy Schoen is vice-president for health policy, re- search, and evaluation at the Commonwealth Fund. Catherine DesRoches is a se- nior research associate at the Harvard School of Public Health. Robin Osborn is assistant vice-president and director of international health policy at the Com- monwealth Fund, and Kinga Zapert is a vice-president of Harris Interactive. International Comparisons 106 May/June 2003 ©2003 Project HOPE?The People-to-People Health Foundation, Inc. ABSTRACT: This article reports on a comparative survey of sicker adults in Australia, Can- ada, New Zealand, the United Kingdom, and the United States. The study finds that despite differences among the health care systems, large proportions of citizens across the five countries report dissatisfaction with their health care system and serious problems includ- ing medical and medication errors, faulty patient-physician communication, and poor care coordination. The most crucial policy implication of these findings is that a focus on a small population of intensive health system users could have the potential to both control costs and improve care. I n their efforts to improve the quality and safety of medical care, the United States and other industrialized countries have increasingly focused on the care experiences of people with chronic care needs or with more acute, in- tensive care needs. These sicker adults are among the most dependent on medical care and vulnerable to variations in its quality and outcomes. Sicker adults also are likely to be at risk for failures to coordinate care and access barriers, including those flowing from efforts to contain rising costs. 1 Focusing on the experiences of sicker adults in the United States, Australia, Canada, New Zealand, and the United Kingdom, the 2002 Commonwealth Fund International Health Policy Survey of Sicker Adults assesses sicker adults? views of these countries? health care systems and priorities for policymakers. The survey captures their recent experiences with care coordination, physician-patient com- munication, medical errors, prescription drug problems, and difficulties getting medical care when needed. c110 International quality agendas. Concerns with quality and patients? experi- ences are high on the policy agendas of these five countries. The study is part of an ongoing effort to promote cross-national learning and collaboration. 2 In the United States, despite health care spending levels that lead the world, recent reports on medical error and system performance issued by the Institute of Medicine have sparked an array of public and private initiatives to improve care and safety. 3 The United Kingdom has made a major investment in quality through an infusion of new public resources derived from tax increases, in an effort to revamp its National Health Service (NHS) to reduce waiting times, improve care, and produce a more patient-centered care system. 4 In Canada, funding increases for Canada?s Medicare since 2000 have followed an earlier decline in the 1990s, and the country is now awaiting a response to recommendations for reforms to address concerns about re- source shortages, waiting lists, and drug coverage. 5 Australia has taken the lead on patient safety, is supporting public policies to enhance private insurance supple- ments of public insurance, and has implemented innovative care models for chroni- cally ill populations in coordinated care trials. 6 New Zealand also is pursuing efforts to improve access and the organization of services within its care system. 7 c110 The survey series. The 2002 survey is the fifth in a series, begun in 1998, of these five English-speaking countries. Survey populations have included the general Common Concerns HEALTH AFFAIRS ~ Volume 22, Number 3 107 public, the elderly, physicians, and now sicker adults. Results are presented at an an- nual meeting attended by the health ministers of each of the five countries. The re- sults are also used by decisionmakers in each country. 8 This survey?s focus on sicker adults highlights the challenges these and other advanced industrialized countries share as they seek to redesign medical care de- livery systems to make care more effective, efficient, safe, and responsive to pa- tients. Survey findings point to common concerns with varying performance, in- dicating opportunities to learn from cross-national initiatives. Survey Methods c110 Screening interviews. The survey screened initial random samples of adults age eighteen or older who met at least one of four criteria: reported their health as fair or poor; reported that they had had serious illness, injury, or disability that re- quired intensive medical care in the past two years; or reported that in the past two years they had undergone major surgery or had been hospitalized for something other than a normal, uncomplicated delivery. Starting with screening interviews with 2,500?2,900 adults in the five countries, this process resulted in a final sample of 750 or more noninstitutionalized ?sicker adults? in each country. This final survey sample represents one-fourth to one-third of those initially contacted (Exhibit 1). The percentages of adults identified who met any one of the four screening criteria were remarkably similar across the five countries, except for the lower proportion of New Zealand adults rating their health as fair or poor. This screening method yielded a study group of sicker adults with either ongoing health care needs or re- cent, more intensive use of the medical care system (Exhibit 1). c110 Survey design. The survey questionnaire was designed by researchers at the Harvard School of Public Health, the Commonwealth Fund, and Harris Interactive, with the advice and review of experts in each country. Except for minor wording changes to reflect terminology differences, the same instrument was used in each country. Harris Interactive and country affiliates conducted the interviews by tele- phone during March?May 2002. Interviews averaged twenty minutes in length. The survey was conducted in English in all countries, with a French option in Canada and a Spanish option in the United States. c110 Sampling error. All surveys are subject to sampling error. The margin of sam- pling error in each country for similarly sick adult groups is approximately plus or minus four percentage points at the 95 percent confidence level or higher. Results were compared between countries using a T test, except where the overall distribu- tion of the variable is shown in the exhibit, when a chi-square test was used. Text and exhibits indicate where differences between countries and groups are signifi- cant. Reading rows from left to right, the exhibits compare each country with those following it, indicating where country pairs differ significantly. 108 May/June 2003 International Comparisons Survey Findings c110 System views. The survey first asked sicker adults about their satisfaction with the health care system overall, with follow-up questions to probe for their main concerns. To assess overall satisfaction, the survey used a standard four-point satis- faction scale borrowed from the Eurobarometer, a general population survey con- ducted in fifteen European countries. 9 Sizable minorities of sicker adults in each country had negative views of their country?s health care system: The proportion saying that they were ?not very? or ?not at all? satisfied ranged from nearly a third in the United Kingdom (31 percent) to more than two out of five adults in New Zea- land (48 percent) and the United States (44 percent) (Exhibit 2). Sicker adults in New Zealand and the United States were significantly more likely than those in the other three countries were to report being dissatisfied. However, in no country did the proportion saying they were very satisfied top 25 percent. c110 Major problems. To understand the most salient concerns among these sicker adults, the study asked respondents to name the two biggest problems facing their respective systems. Answers varied markedly across countries (Exhibit 2). Cost and inadequate coverage concerns led the U.S. list. It is important to note that respon- Common Concerns HEALTH AFFAIRS ~ Volume 22, Number 3 109 EXHIBIT 1 Screening Questions And Final Sample Characteristics, Commonwealth Survey Of Sicker Adults In Five Countries, 2002 AUS CAN NZ UK US Number of adults initially contacted for screening 2,631 2,489 2,912 2,264 2,629 Results of four survey screening questions Rate health fair or poor Had serious or chronic illness, injury, or disability that required a lot of medical care in past 2 years Hospitalized, other than normal delivery, past 2 years Major surgery in past 2 years 17% a,b 18 17 a,b,c,d 8 15% a,b 18 11 b 7 9% b,c 16 13 6 21% c 17 14 6 16% 17 12 8 Percent answering yes to any of the screening questions 32 a,b 30 a,b 26 b,c 33 c 29 Final survey sample of sicker adults 844 750 750 750 755 Health and recent medical care use among sample Rate health as fair or poor 51% a,b 50% a,b 38% b,c 62% c 55% In past 2 years Had serious or chronic illness, injury, or disability that required a lot of medical care Hospitalized for other than normal pregnancy Major surgery Has ongoing chronic illness 54 a,d 58 b,c,d 26 b 63 a 60 b 42 a 25 b,c 62 a 65 b,c 58 b,c 28 b 70 50 c 45 19 c 65 57 45 30 66 SOURCE: Commonwealth Fund/Harvard/Harris Interactive, 2002 Commonwealth Fund International Health Policy Survey of Sicker Adults. NOTE: Reading from left to right starting with Australia, the letter indicates that the country differs from countries to the right at p < .05. a Different from New Zealand. b Different from United Kingdom. c Different from United States. d Different from Canada. dents may mean different things when they say that cost is the biggest problem: costs to themselves, which could be either the high cost of insurance or gaps in cov- erage, or costs to the health care system overall. In the other four countries, shortages, waiting times, and inadequate govern- ment funding led the list of top concerns, with varying emphasis. More than half of sicker Canadians named shortages of health professionals or hospital beds as the leading problem, followed by waiting times. New Zealanders and Britons named waiting times most frequently. In Australia, shortages and waiting times tied as the two top problems. c110 Recommendations. When asked what is the single most important thing that government could do to improve care, respondents in all but the United States were most likely to call for increased public spending on health care, with about a third endorsing this action. Other responses in these countries indicated support for increased investment in the supply of health professionals or hospitals. Notably, 110 May/June 2003 International Comparisons EXHIBIT 2 Health System Views Among Sicker Adults In Five Countries, 2002 AUS CAN NZ UK US Satisfaction with health system Very satisfied Fairly satisfied Not very satisfied Not at all satisfied Sum of ?not very? or ?not at all? satisfied 15% a,b 48 21 14 35 c,d 21% c 41 23 13 36 b,c,d 14% b 36 32 16 48 b 25% d 41 21 10 31 d 18% 36 25 19 44 Two biggest problems with health care system High cost of health care Inadequate coverage of services Shortages of health professionals/hospital beds Waiting times Inadequate government funding 19 a,b,d 9 d 31 a,c,d 31 b,c,d 20 a,b 13 8 d 54 b,c,d 27 b,c,d 16 b,c 21 6 d 20 b,d 41 d 23 d 6 8 d 33 d 39 d 24 d 48 25 5 3 1 Single most important thing government can do to improve health care Spend more money Reduce waste/reduce fraud/allocate resources better Increase number of health professionals/hospitals Reduce costs Improve coverage of services/people Reduce waiting times 30 c,d 7 b 14 a,c,d 5 d 4 d 6 d 32 d 6 b 19 c,d 1 d 3 d 3 34 b,d 10 b 6 b,d 5 d 5 d 9 d 30 d 17 d 17 d 3 d 1 d 9 d 4 6 2 16 21 ? e Change in quality of care, past two years Worse than two years ago About the same as two years ago Better than two years ago 15 a,c 70 12 24 b,c 64 7 11 69 15 13 67 13 13 70 15 SOURCE: Commonwealth Fund/Harvard/Harris Interactive, 2002 Commonwealth Fund International Health Policy Survey of Sicker Adults. NOTES: See Exhibit 1 for unweighted sample sizes. Reading from left to right starting with Australia, the letter indicates that the country differs from countries to the right at p < .05. Responses to questions about satisfaction and health system problems do not add up to 100 percent because of multiple responses. a Different from Canada. b Different from United Kingdom. c Different from New Zealand. d Different from United States. e Not available. these concerns were echoed by physicians in these four countries when surveyed in 2000, with more than half saying that there were too few hospital beds and more than 30 percent, that there were too few specialists. 10 U.S. opinions were more divided. The single policy action endorsed most fre- quently by U.S. respondents was to improve coverage of services or people. How- ever, only one-fifth named this policy step; an equal proportion said that they were ?not sure? what the government should do (data not shown). c110 Quality of care. The principal focus of the survey was quality of care. To assess views on the direction of recent changes in the performance of the health care sys- tem, the survey asked respondents whether they thought that the quality of care they had received had gotten better or worse over the past two years or had re- mained the same. Two-thirds or more in all five countries saw little recent change. However, Canadians were significantly more likely than those in the other four countries were to report that quality had gotten worse (Exhibit 2). With the excep- tion of Canada, similar shares of adults reported declines or improvements in the quality of care they had received. c110 Coordination of care. Patients with chronic health problems or recent hospi- tal or surgical experiences are at risk for breakdowns in the coordination of their care, absent teams or well-organized care systems. A sizable majority of respondents had been cared for by three or more physicians during the past two years (Exhibit 3). Overall, their experiences indicate frequent problems with coordination of care. As these patients moved through the care system, half reported that they found themselves repeating their health story to multiple health professionals. One- quarter of sicker adults in the United States reported a time in the past two years when medical records or test results did not reach their doctor?s office in time for their appointment, a rate similar to that in the United Kingdom. One of five U.S. and Canadian respondents reported that they had been sent for duplicate tests or procedures by different professionals, rates significantly higher than in Australia or the United Kingdom. Although the percentage of adults reporting coordination failures varied among countries, the frequency of such problems in all five countries indicates common experiences with duplication of effort or delays. These experiences may also fuel the public?s perception of waste in the health system captured in Exhibit 2. Patients also encountered conflicting information from different physicians and health professionals. One-fifth to one-quarter of respondents in each of the five countries reported that they received conflicting information about their care from different health professionals in the past two years. Care coordination problems were particularly frequent among people seeing multiple physicians. Those seeing three or more physicians were more than twice as likely as those seeing fewer physicians were to report receiving conflicting in- formation. Similarly, those seeing three or more physicians were about twice as likely to report duplicate tests by different health care professionals and delays in Common Concerns HEALTH AFFAIRS ~ Volume 22, Number 3 111 care when medical records or tests did not reach the office in time in all countries except Australia, which repeated the pattern but to a lesser degree. c110 Pharmaceuticals. The sharp rise in drug spending and trends toward reliance on multiple medications has increased the importance of patient-doctor communi- cation about prescription drugs. 11 The survey revealed heavy reliance on medications among sicker adults: Two-thirds or more of respondents in each country said that they rely on prescription medications on a regular basis (Exhibit 4). U.S. adults were significantly more likely to report that they regularly take four or more pre- scription drugs than were adults in any of the other four countries. Prescription drugs can have serious side effects and interact with other medica- tions. Yet despite the high proportions of adults who were taking multiple pre- scription drugs, 30 percent (U.S.) to 46 percent (U.K.) of respondents said that their physician had not reviewed and discussed all of the medications with them in the past two years. Restricting the survey sample to those taking medications 112 May/June 2003 International Comparisons EXHIBIT 3 Multiple Physicians And Care Coordination Among Sicker Adults In Five Countries, 2002 AUS CAN NZ UK US Number of different doctors and other health professionals seen in past 2 years One Two Three Four Five or more 13% 25 19 14 26 16% 23 17 13 27 13% 23 19 10 30 16% 18 19 12 30 13% 19 22 14 29 Care coordination experiences in past 2 years Had to tell same story to multiple health professionals Records/results did not reach doctors office in time for appt. Sent for duplicate tests/procedures by different health professionals Received conflicting information from different doctors or health professionals 49 a 14 a,b,c 13 a,b 23 50 a 19 a 20 c 23 47 a 16 a,c 17 a,c 24 c 49 a 23 13 a 19 a 57 25 22 26 Percent experiencing the care coordination problems, by number of doctors seen in the past 2 years Received conflicting information 1?2 doctors 3 or more doctors 13 31 d 11 32 d 13 32 d 9 26 d 13 34 d Tests did not reach office in time for appointment 1?2 doctors 3 or more doctors 11 16 12 25 d 9 21 d 14 29 d 15 30 d Sent for duplicate tests 1?2 doctors 3 or more doctors 10 16 d 9 28 d 9 20 d 6 17 d 11 28 d SOURCE: Commonwealth Fund/Harvard/Harris Interactive, 2002 Commonwealth Fund International Health Policy Survey of Sicker Adults. NOTES: See Exhibit 1 for unweighted sample sizes. Reading from left to right starting with Australia, the letter indicates that the country differs from countries to the right at p < .05. a Different from United States. b Different from Canada. c Different from United Kingdom. d Difference between 1?2 doctors and 3 or more doctors significant at p < .05. regularly, these rates still remained high: More than one of five said that their med- ications had not been reviewed in the past two years. Sicker adults in the United Kingdom were significantly more likely to report that their physician had not re- viewed their medications than were adults in the other four countries. The survey found that medication-related side effects, which pose problems for some sicker adults, are not always discussed between doctors and patients. Among all respondents, including those not currently taking medications regu- larly, 17?19 percent said that they stopped taking a prescription medication with- out a doctor?s advice because of side effects, and about one of ten reported that they had experienced serious side effects that the doctor did not tell them about. Few reported having difficulty understanding the instructions for a medication. Compared with adults in the other four countries, U.S. adults were the most likely to say that they skipped doses of medications to make them last longer, a problem found to be widespread among the U.S. elderly in a recent eight-state study. 12 c110 Patient safety: medication and medical errors. Medical errors have become a highly visible quality-of-care issue, both in the media and in the professional litera- ture of many countries. To compare safety issues across five different health care sys- tems, the survey asked sick patients about medication errors and a more general question about any medical mistakes or errors (Exhibit 5). About one in ten sick pa- tients in each country reported a time in the past two years when they were given the wrong medication or wrong dose by a doctor, hospital, or pharmacist. Reports of Common Concerns HEALTH AFFAIRS ~ Volume 22, Number 3 113 EXHIBIT 4 Prescription Drug Use Among Sicker Adults In Five Countries, 2002 AUS CAN NZ UK US Use prescription medicine on a regular basis 66% 64% a 65% a 67% 71% Number of prescriptions used regularly b None One Two to three Four or more 34 a 17 26 23 36 a 13 26 23 35 a 14 25 25 33 a 16 29 22 29 11 24 36 Review of medications by physician relied on the most In the past 2 years, doctor has not reviewed and discussed all of the medications taken 41 a,c,d 37 a,d 34 a,d 46 a 30 Problems taking prescription medications Stopped taking without doctor?s advice because of side effects Taking medication but serious side effects doctor didn?t tell about Skip doses to make the medication last longer Taking but difficulty understanding medication instructions 15 a 11 9 a 3 17 9 8 a 2 16 9 7 a 2 16 11 6 a 3 19 9 16 3 SOURCE: Commonwealth Fund/Harvard/Harris Interactive, 2002 Commonwealth Fund International Health Policy Survey of Sicker Adults. NOTES: See Exhibit 1 for unweighted sample sizes. Reading from left to right starting with Australia, the letter indicates that the country differs from countries to the right at p < .05. a Different from United States. b Significant statistical difference in overall distribution at p < .05. c Different from New Zealand. d Different from United Kingdom. medical mistakes were more frequent. Overall, at least one-fourth of sicker adults in Australia, Canada, New Zealand, and the United States and 18 percent in the United Kingdom reported either a medication or medical mistake in the past two years. Among those who reported either type of error, the majority in each of the five countries said that the mistake caused serious health problems. As a percentage of all respondents, 18 percent of sicker adults in the United States and New Zealand and similar proportions in Australia and Canada reported that a mistake or error had caused a serious prob- lem in the past two years (Exhibit 5). The U.K. had the lowest reported rate of se- rious medical errors. Other studies have found that the number of doctors seen, the number of pre- scription drugs taken, and the probability of an adverse event are correlated. 13 Similar to the earlier study, this survey finds that the incidence of reported medi- cal errors increased among those who saw three or more physicians. Such patients reported errors at about twice the rate of those seeing only one or two physicians. Rates of medication errors also rose with the number of medications taken reg- ularly. At least one of seven respondents who were taking four or more medica- tions regularly said that they had been given the wrong dose or wrong medication in the past two years (Exhibit 5). c110 Doctor-patient communication. For patients with ongoing care needs, mak- ing care more patient-centered by involving patients in care decisions and clearly communicating treatment goals have the potential of improving care. Yet in the sur- 114 May/June 2003 International Comparisons EXHIBIT 5 Medication And Medical Errors Among Sicker Adults In Five Countries, 2002 AUS CAN NZ UK US Mistake made in past 2 years Believed a medical mistake was made in treatment or care Given the wrong medication or wrong dose by a doctor, hospital, or pharmacist Either type of error was made 19% a,b 11 23 a,b 20% a 11 25 a 18% a,b 13 23 a,b 13% a,b 10 18 b 23% 12 28 Mistake caused serious health problems As percent of those who experienced a medical error As percent of all respondents 55 13 a 60 15 a 60 18 a 51 b 9 b 63 18 Percent experiencing a medication or medical error by number of doctors seen in past 2 years 1?2 doctors 3 or more doctors 16 28 c 16 32 c 15 29 c 13 21 c 19 34 c Percent of sicker adults taking 4 or more medications regularly who were given wrong dose or wrong medication in past 2 years 15 14 16 17 16 SOURCE: Commonwealth Fund/Harvard/Harris Interactive, 2002 Commonwealth Fund International Health Policy Survey of Sicker Adults. NOTES: See Exhibit 1 for unweighted sample sizes. Reading from left to right starting with Australia, the letter indicates that the country differs from countries to the right at p < .05. a Different from United Kingdom. b Different from United States. c Difference between 1?2 doctors and 3 or more doctors significant at p < .05. vey at least half of respondents reported that their regular physician does not ask for their ideas or opinions about treatment and care (Exhibit 6). Sizable proportions of respondents said that their physicians do not make clear the specific goals for treat- ment. Moreover, between 28 percent (Canada) and 43 percent (U.K.) of sicker adults said that their doctor does not keep them motivated to do the things they need to do. On all measures in this series, sicker adults in the United Kingdom were significantly more likely than those in the other four countries were to report that these conversations were not taking place. The survey also found that one in five respondents or more in each country re- ported a time when they had left their doctor?s office without getting important questions answered. Rates were highest among U.S. adults (Exhibit 6). Patients also left without following their doctor?s advice or treatment plans. One-fifth of sicker adults in the United Kingdom; about three in ten in Australia, Canada, and Common Concerns HEALTH AFFAIRS ~ Volume 22, Number 3 115 EXHIBIT 6 Physician Communication And Patient-Physician Interactions Among Sicker Adults In Five Countries, 2002 AUS CAN NZ UK US Patient-physician communication Regular doctor or health professional does not Make clear the specific goals for treatment Help understand what needs to be done for health Ask for ideas and opinions about treatment and care Keep patient motivated to do what needs to be done 23% a 12 a,c 51 a 29 a 21% a 14 a 49 a 28 a,c 25% a,b 17 a 47 a 34 a 38% b 26 b 67 b 43 b 20% 14 47 30 In the past 2 years doctor has not Provided advice on weight, nutrition, exercise, smoking, drinking Discussed emotional burden of coping with condition 42 a,b 54 a 38 a,b 55 a 40 a,b 54 a 49 b 66 b 33 51 Getting questions answered and following physician advice Left doctor?s office without getting important questions answered Time in past 2 years when did not follow doctor?s advice or treatment plan 21 b 31 a,b 25 a,b 31 a,b 20 b 27 a,b 19 a,b 21 b 31 39 Reasons why did not follow doctor?s advice or treatment plan Disagreed with what doctor recommended It was too difficult It costs too much Didn?t understand treatment instructions 37 b 31 21 a 7 38 b 35 24 a 5 35 b 30 25 a 6 34 35 7 6 25 33 28 a 7 Rating of physician care Percent rated doctor excellent or very good on How well he or she diagnosed problem Spending enough time Being accessible by phone or in person Listening carefully to health concerns Treating with dignity and respect 67 a,b,d 64 a,b,c,d 57 a,b,c,d 72 a,b,d 79 a,b,d 62 a,c 57 c 52 c 66 c 75 c 68 a,b 72 a,b 68 a,b 76 a,b 83 a,b 57 55 50 65 74 58 52 51 62 71 SOURCE: Commonwealth Fund/Harvard/Harris Interactive, 2002 Commonwealth Fund International Health Policy Survey of Sicker Adults. NOTES: See Exhibit 1 for unweighted sample sizes. Reading from left to right starting with Australia, the letter indicates that the country differs from countries to the right at p < .05. a Different from United Kingdom. b Different from United States. c Different from New Zealand. d Different from Canada. New Zealand; and nearly two in five in the United States reported a time when they had not followed their doctor?s advice or treatment plan during the past two years (Exhibit 6). When asked for reasons why they did not adhere to advice, siz- able proportions in all five countries said that they disagreed with what the doctor recommended or that recommended care was too difficult to follow. In Australia, Canada, New Zealand, and the United States, costs were also a leading reason why patients did not follow their physician?s advice. Considerable emotional strain often accompanies serious or chronic illnesses. Yet the majority of respondents in the five countries said that their regular doctor had not discussed the emotional burden of coping with their illness in the past two years (Exhibit 6). British respondents were the least likely to report emo- tional support from their physician. Although physicians are doing marginally better at talking about exercise, diet, weight, smoking, and other lifestyle issues, one-third or more of adults in each country reported no discussion of these issues in the past two years. Again, British respondents were the most likely to report that these conversations had not taken place. Despite evidence of communication failures, the majority of sicker adults in all five countries rated their physician highly (excellent or very good) on questions about care relationships: listening to their health concerns, treating them with dignity and respect, being accessible, spending enough time with them, or diag- nosing their problems correctly (Exhibit 6). Ratings of ?excellent? or ?very good? tended to be highest in New Zealand, with Australia a close second and signifi- cantly lower ratings in both the United States and the United Kingdom. Sicker U.S. and U.K. adults were also more likely than their counterparts were to give their physician a negative (fair or poor) rating on diagnosis. In New Zealand, high ratings of personal physicians coexisted with high levels of dissatisfaction with the health care system overall (Exhibit 1). Analysis of the interaction of care experiences and system views in New Zealand indicate that New Zealanders who were dissatisfied with the care system were significantly more likely (typically twice the rate or more of those who were satisfied with the system) to have experienced medication or medical errors, care coordination problems, and medication side effects and to have encountered difficulties access- ing the care system, including waits or access problems related to cost (data not shown). This pattern also held in the other four countries. c110 Access to care and waiting times. The five countries vary greatly in insur- ance systems; availability of resources such as physicians, specialists, and hospitals; and recent histories of public investment in their health care systems. 14 Sicker adults? responses to questions about access to care tended to reflect these variations in countries? insurance and resources. These findings repeat patterns observed in the 1998 and 2001 five-nation surveys. 15 Seeing specialists. Half of Canadians said that it was difficult to see a specialist when needed, citing waits for appointments as the dominant reason (Exhibit 7). 116 May/June 2003 International Comparisons Difficulty in seeing specialists was also reported by more than one in three respon- dents in Australia, New Zealand, and the United Kingdom, with waiting times listed as the primary reason. Despite the fact that the United States has more prac- ticing specialists per 1,000 people than the other four countries, two of five U.S. re- spondents also reported difficulties seeing a specialist when needed. U.S. adults, along with adults in New Zealand and Australia, cited costs as one of the main rea- Common Concerns HEALTH AFFAIRS ~ Volume 22, Number 3 117 EXHIBIT 7 Waiting Times And Access Concerns Among Sicker Adults In Five Countries, 2002 AUS CAN NZ UK US Difficulty seeing specialist when needed Very difficult to see specialist a Somewhat difficult to see specialist Sum of ?very? and ?somewhat? difficult to see specialist 17% 24 41 b 24% 29 53 c,d,e 12% 24 36 17% 21 38 15% 24 40 Reasons difficult to see specialist (base: those saying difficult) Having to wait for appointment, long waiting times for type of care Unable to afford/lack of (private) insurance Being denied referral or having to wait for referral Facilities or service not available locally or lack of doctors available 74 b,c,e 17 b,c,d 2 b,c,d,e 18 b,e 86 c,d,e 3 c,e 10 d,e 24 c,d,e 61 d,e 23 d,e 10 d,e 19 d,e 75 e 5 e 6 e 15 40 17 31 13 Problem in past 2 years Long waits to be admitted to hospital Big problem a Small problem Not a problem 20 b,e 11 59 28 c,d,e 16 49 21 e 11 63 19 e 12 51 13 13 70 Long waits to get appointment with regular doctor Big problem a Small problem Not a problem 17 b,c,d 18 64 24 c,e 21 53 5 d,e 10 84 21 e 18 57 14 24 61 Doctors not spending enough time Big problem a Small problem Not a problem 13 b,d,e 15 71 20 c,d 20 58 9 d,e 13 78 12 e 14 69 20 26 54 Delay of scheduled surgery or other medical procedure because of cancellation Big problem a Small problem Not a problem 10 b,e 7 73 16 c,d,e 11 66 9 e 10 75 10 e 8 62 5 9 79 Access problems due to cost Did not fill a prescription Did not get medical care Did not get test, treatment, or follow-up Did not get dental care 23 d,e 16 c,d,e 16 b,d,e 44 b,d 19 d,e 9 c,d,e 10 c,d,e 35 c,d,e 20 d,e 26 d 15 d,e 47 d,e 10 e 4 e 5 e 21 e 35 28 26 40 Cost of overall medical care, including any services needed to cope with chronic illness, is a major burden 20 d 18 d,e 21 d 9 e 23 SOURCE: Commonwealth Fund/Harvard/Harris Interactive, 2002 Commonwealth Fund International Health Policy Survey of Sicker Adults. NOTES: See Exhibit 1 for unweighted sample sizes. Reading from left to right starting with Australia, the letter indicates that the country differs from countries to the right at p < .05. a Significant statistical difference in overall distribution at p < .05. b Different from Canada. c Different from New Zealand. d Different from United Kingdom. e Different from United States. sons for these difficulties. Costs of specialty care were not an issue in Canada or the United Kingdom. It was only in the United States that sicker adults cited refer- ral denials or delays as a leading reason for difficulties seeing a specialist. Hospital access. The survey also asked whether waits to be admitted into the hos- pital or to see a doctor or delays in surgery because of cancellation had posed prob- lems in the past two years. A 2001 survey in the same five countries found waits of four months or more for elective admissions to hospitals in the United Kingdom and almost no waiting times in the United States. 16 When sicker adults were asked the more subjective question in 2002 regarding whether waits had been a problem, the gap between countries narrowed. However, U.S. adults remained least likely to report problems waiting to enter a hospital. The share of sicker adults saying that waiting for hospital admission was a ?big? problem was highest in Canada (Exhibit 7). Restricting the base to adults hospitalized in the past two years, Canadians remained the most likely (32 percent) to report big problems waiting for admission. Physician access. Regarding waits to see a physician, respondents in New Zealand were the least likely to report big problems, and Canadian, U.K., and U.S. respon- dents were the most likely (Exhibit 7). These responses may reflect expectations as well as actual differences in waiting times. The 2001 survey in the five countries asked about the number of days adults had to wait to see a doctor when sick and found that the majority of patients in New Zealand and Australia (69 percent and 62 percent) saw a doctor within one day when sick, with longer waits in the other three countries. Delayed surgery. Compared to waits for hospital admissions or doctor appoint- ments, comparatively few sicker adults reported big problems with delay of sched- uled surgery or other medical procedures because of cancellation (Exhibit 7). Cost barriers. Access problems are also related to cost. Sicker adults in the United States were most likely to say that they had not filled a prescription or followed up on recommended tests or treatment because of costs, and U.K. respondents were the least likely to report these problems (Exhibit 7). New Zealanders and U.S. adults were the most likely to say that they did not get needed medical care be- cause of cost. At least one of four sicker U.S. adults reported forgoing a needed medical service because of cost. Sicker adults in the United Kingdom were signifi- cantly less likely to report forgoing needed care because of costs than were adults in the other four countries. In general, access questions related to costs were among the only measures in the survey on which care experiences varied significantly by income. On questions about coordination of care, medical or medication errors, or communication with 118 May/June 2003 International Comparisons ?A focus on a small population of intensive users could have the potential for improving care for all citizens.? physicians, the study found few or no significant differences by income in the five countries (data not shown). Commonalities And Differences The small proportions of all five countries? populations that are sicker account for the majority of health care expenditures in any given year. 17 This survey indi- cates that these sicker adults are likely to be at high risk for deficiencies in care co- ordination, communication failures, and medical care errors. As countries seek to improve care quality, effectiveness, and safety, the survey finds notable areas of common concern as well as varying performance across the five nations. A limitation of this cross-national study is that we are unable to assess how much of these differences can be attributed to variations in cultural expectations. Despite this limitation, the similarities in problems of coordination, communica- tion, medication problems, and errors are striking. In all five countries sicker adults? experiences indicate that failure to coordinate care can result in duplicate tests, delays in care, wasted patient and medical staff time, and conflicting information. Deficiencies in patient-doctor communication can compound such concerns because of failure to involve patients in decision making. The survey indicates frequent failures of physicians during patient visits to exchange information and answer questions, to discuss care goals and options, and to review medication regimens. Although the United Kingdom stands out in this regard, reports from all five countries suggest that each country is missing op- portunities to improve care and decrease the possibility of errors. c110 United Kingdom. Access concerns also exist in all five countries, but sources and types of barriers vary widely. These system differences influence public opinion and expectations of policy leaders. Sicker British adults reported problems with waiting times and other nonfinancial barriers to care, but they were the most satis- fied with their health care system. c110 Canada. Shortages of physicians or other resources and waiting times also top the list of public concerns in Australia, Canada, and New Zealand. Sicker Canadians stand out as the most likely to report difficulties in seeing specialists, and a substan- tial proportion perceived that quality of care had declined in recent years. Levels of discontent in Canada likely reflect high expectations and recent memories of more ready access to care. Until the early 1990s Canadians expressed the greatest satisfac- tion with their national health care system, but ratings plummeted with restrictions in national expenditures for health care during the mid-1990s. 18 The sharp increase in the level of discontent in Canada underscores the high value the public places on their health care system. Canadians are now awaiting a policy response to the report from the Royal Commission on the Future of Health Care in Canada. c110 New Zealand. New Zealand adults expressed some of the highest levels of dis- satisfaction with their system overall, citing concerns with waiting times for hospi- tal care and endorsing increased public funding for health care. New Zealand has Common Concerns HEALTH AFFAIRS ~ Volume 22, Number 3 119 undergone a series of major health reforms over the past decade. 19 The public re- sponse to system changes might account for some of this discontent. Concerns also might reflect capacity constraints. New Zealand?s health spending as a percentage of gross domestic product (GDP) remains below the Organization for Economic Co- operation and Development (OECD) median. 20 New Zealanders, however, give their physicians high ratings and report few problems with accessing doctor?s offices. c110 Australia. Sicker adults in Australia also reported long waiting times and shortages as their top problems. However, Australia was generally in the middle in terms of the range of problems faced by the five countries. It was neither the best nor the worst on any measure. c110 United States. Sicker U.S. adults were most likely to be concerned about costs and coverage and to report access barriers due to costs. They stand out for forgoing medical care and not getting recommended follow-up treatment because of costs, including skipping medications. This unique exposure to financial burdens when sick reflects basic U.S. insurance patterns. Although U.S. per capita spending and the percentage of GDP spent on health leads the world, forty-one million Americans remain uninsured, and those with insurance tend to face higher out-of-pocket costs than do insured populations in other countries. 21 It is also surprising, given the much higher level of spending, that the United States does not rank higher on most mea- sures in this survey compared to the other five countries. Notably, it ranked poorly on care coordination, medical errors, overall rating of doctors, and getting questions answered. The high rates of duplicate tests and coordination failures in the United States may contribute to higher costs as well as negative patient care experiences. O urfindingsindicate that as countries seek to redesign their care sys- tems, interventions that target patients who see multiple physicians or de- pend on multiple medications could be particularly effective. Tracking systems, shared electronic medical records, or electronic prescribing of medica- tions offer potential payoffs. Better monitoring of prescriptions could help to pre- vent medication-related adverse interactions, side effects, and prescription errors and could work toward controlling high pharmaceutical costs. Because sicker U.S. adults are more likely than their counterparts in the other four countries are to see multiple physicians and to be taking multiple medications, efforts to improve care coordination in the United States could be particularly fruitful. Sicker patients, especially those seeing multiple doctors and taking multiple medications, are bellwethers for how well a health care system is working. The most important policy implication of this study is that a focus on a small popula- tion of intensive users of the health care system could have the potential for con- trolling costs and improving care for all citizens. 120 May/June 2003 International Comparisons An earlier version of this work was presented at the Commonwealth Fund?s international symposium, Reconciling Rising Health Care Costs and Getting Value for Money, 23?25 October 2002, in Washington, D.C. The authors thank Karen Davis and three anonymous reviewers for many thoughtful comments that improved this paper. The views expressed are those of the authors and should not be attributed to the Commonwealth Fund or its directors or officers. NOTES 1. G.F. Anderson and J.R. Knickman, ?Changing the Chronic Care System to Meet People?s Needs,? Health Af- fairs (Nov/Dec 2001): 146?160. 2. Commonwealth Fund, 2002 International Symposium on Health Care Policy, ?Reconciling Rising Health Care Costs and Getting Value for Money,? Washington, D.C., 23?25 October 2002. 3. Institute of Medicine, Crossing the Quality Chasm: A New Health System for the Twenty-first Century (Washington: National Academies Press, 2001); and L.T. Kohn, J.M. Corrigan, and M.S. Donaldson, eds., To Err Is Human: Building a Safer Health Care System (Washington: National Academies Press, 1999). 4. ?Budget 2000 Government States: Reform Is Price of NHS Cash Boost,? Financial Times, 22 March 2000, 31. 5. R.J. Romanow, Building on Values: The Future of Health Care in Canada?Final Report (Ottawa: Canadian Govern- ment Publishing, November 2002). 6. S. Willcox, ?Promoting Private Health Insurance in Australia,? Health Affairs (May/June 2001): 152?161; and Commonwealth Department of Health and Ageing Care, ?The Australian Coordinated Care Trials: Sum- mary of the Final Technical National Evaluation Report of the First Round of Trials? (Canberra: Depart- ment of Health and Ageing, 2001). 7. N. Devlin, A. Maynard, and N. Mays, ?New Zealand?s New Health Sector Reforms: Back to the Future?? British Medical Journal (12 May 2001): 1171?1174. 8. For the other surveys, see K. Donelan et al., ?The Cost of Health System Change: Public Discontent in Five Nations,? Health Affairs (May/June 1999): 206?216; C. Schoen et al., ?Health Insurance Markets and Income Inequality: Findings from an International Health Policy Survey,? Health Policy 51, no. 2 (2000): 67?85; K. Donelan et al., ?The Elderly in Five Nations: The Importance of Universal Coverage,? Health Affairs (May/ June 2000): 226?235; R.J. Blendon et al., ?Physicians? Views on Quality of Care: A Five-Country Compari- son,? Health Affairs (May/June 2001): 233?243; and R.J. Blendon et al., ?Inequities in Health Care: A Five-Country Survey,? Health Affairs (May/June 2002): 182?191. 9. Eurobarometer public opinion polls are conducted on behalf of the European Commission. See German Social Science Infrastructure Service, ?The Eurobarometer Survey Series,? 18 November 2002, www.gesis. org/en/data_service/eurobarometer (13 February 2003). 10. Blendon et al., ?Physicians? Views on Quality of Care.? 11. G. Anderson, V. Petrosyan, and P.S. Hussey, Multinational Comparisons of Health Systems Data, 2002 (New York: Commonwealth Fund, October 2002). 12. D. Safran et al., ?Prescription Drug Coverage and Seniors: How Well Are States Closing the Gap?? 31 July 2002, www.healthaffairs.org/WebExclusives/Safran_Web_Excl_073102.htm (10 February 2003). 13. J.L. Wolff, B. Starfield, and G. Anderson, ?Prevalence, Expenditures, and Complications of Multiple Chronic Conditions in the Elderly,? Archives of Internal Medicine 162, no. 20 (2002): 2269?2276. 14. Anderson et al., Multinational Comparisons. 15. Donelan et al., ?The Cost of Health System Change?; Schoen et al., ?Health Insurance Markets?; and Blendon et al., ?Inequities in Health Care.? 16. In the United States only 5 percent of those with a hospitalization waited four months or more, compared with 23 percent in Australia, 26 percent in New Zealand, 27 percent in Canada, and 38 percent in the United Kingdom. Blendon et al., ?Inequities in Health Care.? 17. B.G. Druss et al., ?Comparing the National Economic Burden of Five Chronic Conditions,? Health Affairs (Nov/Dec 2001): 233?241. 18. Donelan et al., ?The Cost of Health System Change.? 19. Blendon et al., ?Inequities in Health Care.? 20. Anderson et al., Multinational Comparisons. 21. Ibid. Common Concerns HEALTH AFFAIRS ~ Volume 22, Number 3 121 Robert J. Blendon, Cathy Schoen, Catherine DesRoches, Robin Osborn, and Kinga Zapert Common Concerns Amid Diverse Systems: Health Care Experiences In Five Countries