37
NKDEP Survey of Primary Care Providers’ Knowledge and Practices Related to Kidney Disease Draft Report from the Baseline Study Date: September 2003 Prepared for: National Institute of Diabetes and Digestive and Kidney Diseases Submitted by: Equals Three Communications, Inc. OMB Control No. 0925:0515, Exp. 1/31/2005

Respondent Characteristics

Embed Size (px)

Citation preview

Page 1: Respondent Characteristics

NKDEP Survey of Primary Care Providers’ Knowledge and Practices Related to Kidney DiseaseDraft Report from the Baseline Study

Date: September 2003

Prepared for: National Institute of Diabetes and Digestive and Kidney Diseases

Submitted by: Equals Three Communications, Inc.

OMB Control No. 0925:0515, Exp. 1/31/2005

Page 2: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

Table of Contents

Introduction to the Project............................................................ 3

Methodology....................................................................................3

Data Collection...................................................................................3

Sample Characteristics.....................................................................5

Data Analysis and Reporting ............................................................6

Summary of Findings ....................................................................7

Risk Factors.................................................................................................7

Diagnosis and Screening............................................................................8

Treatment Recommendations..................................................................11

Communicating about CKD and its Risk Factors...................................14

Appendix: Survey Instrument...............................................................18

: Introduction to the Project

The National Kidney Disease Education Program (NKDEP) is a pioneering program to reduce the economic, social and human burden of chronic kidney disease (CKD) and kidney failure by encouraging prevention, early detection and treatment of CKD among high-risk individuals and early CKD patients. Prior to launching a national health education campaign, NKDEP is conducting pilot-site interventions in four locations to refine and test campaign strategy. The pilot intervention targets African Americans adults – a population at higher risk for kidney disease – and their primary care providers (PCPs). Four pilot sites1 were selected based on the large population of African Americans in each location and the existing availability of partnership networks and resources. The program, begun in April 2003, targets (1) African Americans at risk for kidney disease, specifically those who have diabetes, hypertension and/or a family history of kidney failure, and (2) primary care providers, specifically family practitioners, general internists, nurse practitioners and physician assistants. A composite control site is comprised of the Memphis, TN, St. Louis, MO and New Orleans, LA metropolitan areas.

In the pilot sites, community-based communication programs are being implemented to educate African American adults (aged 30 and older) to assess their risk status, to persuade those who are at risk to get tested regularly for CKD and take steps to prevent CKD, and to motivate those who have CKD to take

1 Cleveland, OH; Baltimore, MD; Atlanta, GA; and Jackson, MS.

2

Page 3: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

steps to slow its progression. The pilot programs are also educating primary care providers in these communities to monitor at-risk patients more effectively, to communicate better with patients regarding CKD, and to combat early-stage CKD more aggressively through tighter glycemic and blood pressure control and appropriate medication use.

This research study was conducted to provide a baseline for measurement of program effects among primary care providers, and to validate the model of program effects that has guided the development of the program. The results of this study and the follow-up survey will also form the basis of the communication strategy for the next phase of the program.

Methodology

Data Collection

A multi-modal survey of 665 primary care providers was conducted between February 20 and April 10, 2003. Lists of PCPs in the pilot site and control site locations were purchased and all potential respondents were approached through a variety of contact methods in order to maximize survey response.2 The original design of the study called for conducting approximately 100 interviews in each of the four pilot site locations and 200 interviews within the composite control, for a total of 600 completed interviews. However, when it was determined that sufficient sample was not available for the Jackson, MS pilot site, the quota from that location was redistributed to the remaining pilot sites, allowing for approximately 133 interviews per pilot site. The final sample size in the pilot sites exceeded the initial goal in order to increase the number of nurse practitioners and physician assistants in the sample. In all, 468 surveys were completed in the pilot sites and 197 in the control site.

2 Data collection was performed by the Attitude Measurement Corporation in Southampton, PA.

3

Page 4: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

The basic data collection method was to pre-screen potential respondents via telephone to ensure their qualification for the research and then fax a survey to qualified recruits. In order to qualify for participation each respondent had to meet the following criteria:

Be a physician, nurse practitioner or physician assistant

Work in a family, general or internal medicine practice

See at least 10 adult patients per week

Work in a solo practice, group practice, HMO, hospital setting or community clinic

Have at least 20% of patients with diabetes and/or hypertension

Completed surveys were returned via fax or mail. In addition to reaching respondents via telephone, the following additional recruiting methods were used:

For those respondents lacking a valid telephone number, an invitation letter was faxed with information about the research and a toll-free number for the PCP to call. Callers were asked the same screening questions and upon qualification, were faxed the survey.

To reach sample size targets, the sample list was expanded, and a letter on NKDEP letterhead and signed by the NKDEP director was mailed out to all PCPs who had not yet responded inviting them to participate in the research by calling a toll-free number. Callers were then pre-screened and qualified recruits were faxed the survey.

To thank them for their participation, each PCP received a $50 honorarium and was offered a summary of the survey results.

Assuming the sampling procedure outlined above produced a random sample of the population of interest, the estimated theoretical standard error associated with the sample estimates obtained (n=665) ranges from .02 (when the population estimate is 50%) to about .008 when the estimate is 95% or 5%. The corresponding sampling margin of error of the population estimates at the 95% confidence level is 4% when the estimate is close to 50% and then declines at the upper and lower ends of the scale.3

In addition to sample size, the quality of a sample is determined by cooperation rate; that is, the proportion of members of the original sample who provide an interview. The overall cooperation rate for this study was 88%, i.e. 88% of the eligible respondents contacted for the survey faxed back the completed survey.4

3 The standard error is used to estimate the sampling margin of error of the estimates (i.e., the probable difference in results between interviewing the entire population Primary Care Providers in the target cities versus taking a scientific sample of the population) that extend 1.96 standard error units around that value (i.e. the 95% confidence level). The standard error is calculated according to the following formula:

P +/- 1.96 * (standard error)4 Cooperation rate was computed using the American Association for Public Opinion Research (AAPOR) guidelines for reporting results of survey. The rate computed here is AAPOR Cooperation Rate 3 (COOP3). COOP3 = Interviews/(Interviews +Partials + Refusals).

4

Page 5: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

Sample Characteristics

Respondents were drawn relatively evenly from each pilot site locations and the composite control sites.

In terms of years of experience, there appeared to be two separate groups, those with 10 or fewer years in practice (47%) and those with over fifteen years (39%).

Most physicians surveyed were board certified (82%). A majority of respondents (62%) were between the ages of 35 and 54. PCP respondents reportedly saw an average of 400 patients a month. On average, about 50% of

these patients have hypertension and/or diabetes. And on average, about 30% of respondent patients are African American.

A vast majority of PCPs work in private practices (82%).

Respondent CharacteristicsLocation n % Position/Specialty n %Pilot Sites 468 70 Physician 465 70 Atlanta, GA 148 22 General/Family Practitioner 190 Baltimore, MD 166 25 Internist 275 Cleveland, OH 154 23 Physician’s Assistant 54 8Control sites 197 30 Nurse Practitioner 146 22

Years in Practice Age5 years or less 185 28 Under 35 116 176 to 10 years 127 19 35 to 44 187 2811-15 years 83 12 45 to 54 223 33More than 15 years 261 40 55 or older 134 20

Board Certification (Physicians only)Yes 378 82No 81 18

Practice CharacteristicsNumber of Patients/Month

n %Percent of Patients with Hypertension and/or Diabetes n %

40 – 200 Patients 163 24 20% - 30% 180 27201 – 300 Patients 98 15 31% - 40% 108 16301 – 400 Patients 163 24 41% - 50% 128 19400 – 500 Patients 135 20 51% - 60% 83 12More than 500 Patients 185 28 61% or more 166 25

Practice Setting Percent of Patients who are African American

Group/Solo Private 546 82 10% or less 163 24Hospital/HMO/Community-based 119 18 11% - 20% 120 18

21% - 30% 101 1531% - 40% 72 1141% - 50% 79 1251% or more 127 19

5

Page 6: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

Data Analysis and Reporting

This report presents the overall sample descriptive statistics and variation by specific demographic, medical and practice-related subgroups. The information has been grouped into topic areas, and the sequence of reporting does not follow the sequence of questions on the survey instrument. The survey instrument is attached for your reference (see Appendix) and data tables provided under separate cover.

Two variables were derived for purposes of analysis. The median number of African American patients (88) was used to split respondents into two categories: those with 88 or fewer African American patients were classified as having a low number of African American patients, and those with 89 or more were classified as having a high number of African American patients. Likewise, the median number of patients who have diabetes and/or hypertension (158) was used to create two categories, those with a low number (158 or fewer) and those with a high number (159 or more).

Mean Number of African American Patients 88

Mean Number of Patients with Diabetes and/or Hypertension

158

Low Number of African American Patients 88 or less

Low Number of Patients with Diabetes and/or Hypertension 158 or less

High Number of African American Patients 89 or more

High Number of Patients with Diabetes and/or Hypertension 159 or more

6

Page 7: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

Summary of Findings

Risk Factors for CKD

The typical practitioner reported they had more active patients with hypertension than with diabetes. However, practitioners generally had large numbers patients with both diseases within their practice. On average, respondents said their practice had 293 active patients with diabetes and 410 active patients with hypertension (Questions 1 and 4, respectively).

o Physicians reported higher numbers of patients with diabetes and hypertension than did non-physicians and internists reported more patients with hypertension than did general practitioners.

o Respondents from solo or group private practices reported higher numbers of patients with hypertension and diabetes than did those from hospital, community or HMO practices.

o Primary care providers with a higher number of African American patients also reported a higher number of patients with diabetes and hypertension.

The majority of respondents correctly believed that diabetes and hypertension greatly increased one’s risk for kidney disease but less than half believed that being African American and less than a quarter believed that a family history of kidney failure did so. Respondents were asked to indicate if they believed risk for CKD was increased not at all, slightly, moderately or greatly by each of several potential risk factors (Question 7). The vast majority said diabetes (90%) and hypertension (76%) greatly increase one’s risk for CKD. Less than one-half (41%) of respondents said African Americans are at greatly increased risk. Approximately one-quarter of respondents indicated that having a family history of kidney failure (24%) or smoking (27%) greatly increased one’s risk.

o More respondents from the control site than the pilot sites believed a family history of kidney failure greatly increases one’s risk of CKD. More control site respondents also said that males or smokers were at greatly increased risk for the disease compared to pilot site respondents.

o Among physicians, non-physicians said African American race, smoking, family history of kidney failure and male gender greatly increased risk more often than did physicians. Internists said hypertension and African American race greatly increased risk more often than did general practitioners.

7

Page 8: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

Diagnosing CKD

Although few practitioners said they always calculate GFR from serum creatinine, nearly half do so part of the time and about a quarter say their lab calculates GFR for them. Specifically, when asked if they estimated GFR from serum creatinine using a calculation, 4% said they always do, 38% said they often do and 10% said they sometimes do (Question 12). Of those who do not regularly calculate GFR, about half say they never use the test (25% of all respondents) and half say a lab calculates GFR for them (22% of all respondents).

o More internists than general practitioners said they calculate GFR always, often or sometimes and more general practitioners than internists said they never calculate GFR. Interestingly, about twice as many general practitioners than internists said their lab calculates GFR for them. More physicians reported calculating GFR always, often or sometimes than did non-physicians.

o Although the percentages are low, twice the number of respondents representing hospital-based, HMO and community practices as those from private practice reported always calculating GFR (7% compared to 3%). Similarly, nearly a third (27%) of those in private practice said they never calculate GFR compared to those in other practice settings.

Respondents were asked to identify the index and specific level at which they would diagnose CKD given the following situation (Question 9): You have a 65-year-old Caucasian woman patient who weighs 50 kg/110 lbs and has had diabetes (HbA1C 8.0) and hypertension (BP 138/90) for several years.

o Providers appear to use creatinine level more often than other indices when diagnosing CKD. Most respondents (88%) indicated a creatinine level to diagnose CKD in the patient scenario. Nearly two-thirds of respondents (61%) indicated a proteinuria level, about one-half (47%) indicated a GFR level and one-quarter (24%) indicated an albumin/creatinine ratio to diagnose CKD in the patient scenario provided. Similarly, among all respondents, just 7% said they do not use creatinine level when diagnosing CKD compared to about a third (29%) of all respondents who said they do not use proteinuria, about half who said they do not use GFR and two-thirds (67%) who said they do not use albumin/creatinine ratio.

More internists than general practitioners indicated a creatinine level to diagnosis CKD in the patient scenario. Similarly, more physicians than non-physicians indicated a creatinine, proteinuria and/or GFR level to diagnosis CKD in the patient scenario .

More group or solo private practice PCPs indicated a creatinine level to diagnosis CKD in the patient scenario than did those in hospital, HMO or community-based practices.

More respondents with higher numbers of active patients with diabetes or hypertension indicated a creatinine level to diagnosis CKD in the patient scenario than those with fewer patients with either condition.

8

Page 9: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

o The diagnosis level indicated by respondents varied widely for each index. The mean diagnosis level for creatinine was 2.37 mg/dl (mode 1.5 mg/dl) and ranged from .2 mg/dl to 300 mg/dl; for proteinuria the mean was 541 mg/24hrs (mode 300 mg/24hrs) and ranged from .03 mg/24hrs to 4000 mg/24 hrs; for GFR the mean was 80 ml/min (mode 60 ml/min) and ranged from 4 ml/min to 3000 ml/min; and for albumin/creatinine ratio the mean was 49 mg/mg (mode 30 mg/mg) and ranged from .02 mg/mg to 300 mg/mg.

o Nearly twice as many internists as general practitioners indicated they would diagnose CKD at creatinine levels below 1.50 mg/dl and twice an many internists as general practitioners indicated they would diagnose CKD in accordance with NKDEP preferences (e.g. at a creatinine level of 1.0 mg/dl or more).

o Respondents in group or solo private practices more often indicated they would require creatinine levels over 1.50 mg/dl than those in hospital-based, HMO or clinic facilities.

More physicians than non-physicians indicated they would diagnose CKD at proteinuria index levels of 300 mg/24 hours or higher. When using GFR, more physicians than non-physicians said they would diagnose CKD at 60 ml/min or higher.

Index Levels used to Diagnose CKD(Percent)

Creatinine level 88 GFR 471.00 mg/dl or less 6 Less than 60 ml/min 191.01 to 1.49 mg/dl 17 60 ml/min 111.50 mg/dl 21 Greater than 60 ml/min 171.51 to 2.00 mg/dl 33 Do not use this index 46Greater than 2.00 mg/dl 11 Don’t know/not reported 7Do not use this index 7Don’t know/not reported 4

Albumin/creatinine ratio 24Proteinuria 61 Less than 30.00 mg/mg 10Less than 300 mg/24 hrs 27 30.00 mg/mg 9300 mg/24 hrs 15 Greater than 30.00 mg/mg 5Greater than 300 mg/24 hrs 19 Do not use this index 67Do not use this index 29 Don’t know/not reported 9

9

Page 10: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

Based on the patient scenario NKDEP would prefer CKD be diagnosed at the levels noted below:

Creatinine 1.00 mg/dl or moreProteinuria 311 mg/24 hrs or moreAlbumin/creatinine ratio 30 mg/mg or moreGFR 60 ml/min or less

To determine the aggressiveness of providers in diagnosing CKD a comparison was made between the providers diagnosis level and NKDEP’s preferred diagnosis level (see table below). Given the wide range of responses for each index it is difficult to draw concrete conclusions about provider’s diagnostic aggressiveness. In most circumstances, however, providers appear to rely on a diagnostic standard that is less aggressive than NKDEP would prefer. The sole exception to this is GFR, when a majority of respondents (41%) indicated a more aggressive standard than preferred by NKDEP.

Aggressiveness of Diagnosing CKD(Percent)

Creatinine level GFRLess aggressiveOver 1.01 mg/dl

93 Less aggressiveLess than 60 ml/min

36

At NKDEP Recommendation1.0 mg/dl

4 At NKDEP Recommendation60 ml/min

23

More aggressiveLess than 1.0 mg/dl

3 More aggressiveOver 60 ml/min

41

Proteinuria Albumin/creatinine ratioLess aggressiveOver 312 mg/24 hrs or more

69 Less aggressive31 mg/mg or more

41

At NKDEP Recommendation311 mg/24 hrs

0 At NKDEP Recommendation30 mg/mg

36

More aggressiveLess than 311 mg/24 hrs

31 More aggressive30 mg/mg or less

23

Treatment Recommendations

Most practitioners appear to take—or at least be aware of—steps that would prevent or slow the progression of CKD. Based on the patient scenario described previously, respondents were asked to indicate which steps they would typically take (Question 10). Nearly all respondents said they would aim for lower blood pressure, aim for tighter glycemic control, measure the lipid profile, monitor kidney function more closely and prescribe ACE inhibitors or angiotensin receptor blockers (ARBs). Nearly all respondents said they would also caution the patient again NSAID use and provider her with educational materials.

o Non-physicians were more likely than physicians to refer the patient to a specialist, and general practitioners were more likely to do so than internists.

o In general, more physicians than non-physicians said they would measure lipid profile, monitor kidney function more closely, prescribe ACE inhibitors or ARBs and caution against NSAID use.

o General practitioners and internists both said they would lower blood pressure, measure lipid profiles, and monitor kidney function more closely. Virtually all internists said they would

10

Page 11: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

also aim for tighter glycemic control, prescribe ACE inhibitors (or ARBs) and caution against NSAID.

o More PCPs with higher numbers of patients with diabetes or hypertension said they would caution patients against NSAID use and refer the patient to a specialist than did respondents with fewer patients with diabetes or hypertension.

Respondents were asked to identify which tests or procedures they regularly recommend to patients with diabetes on an annual basis and the percentage of these patients who receive each procedure (Question 2). As shown in the table below, nearly all respondents indicated they conduct hemoglobin A1c, serum creatinine, lipid profiles and foot exams on patients with diabetes annually. Most respondents also said that nearly all of their patients with diabetes receive each of these procedures. Approximately two-thirds of PCPs (63%) use a dipstick for microalbuminuria and do so with a majority (78%) of their patients with diabetes. Not surprisingly, fewer respondents said they used the 24-hour urine tests and they used them with fewer of their patients annually. About two-thirds of respondents said they use either test (68% for the 24 hour urine for albumin or protein, and 69% for the 24 hour urine for creatinine clearance) and only about one-third of their patients undergo such procedures annually (33% and 28%, respectively).

Tests Routinely Used for Patients with Diabetes(Percent)

ProcedurePCPs

that use it Patients that get

annuallyHemoglobin A1c 99 97Serum Creatinine 99 95Lipid Profile 99 93Foot Exams 97 88Dipstick for Proteinuria 87 84Dipstick for Microalbuminuria 63 78Dilated Retinal Exam 73 76Spot Urine Albumin/Creatinine Ratio 64 6824 hour Urine for Albumin or Protein 68 3324 hour Urine for Creatinine Clearance 69 28

o Compared to physicians, non-physicians estimated that a higher percentage of their patients with diabetes received a 24-hour urine test for creatinine clearance and dipstick tests for proteinuria/urinalysis and for microalbuminuria. However, non-physicians also estimated that a lower percentage of their patients with diabetes received a serum creatinine test or lipid profile than did physicians.

o Internists estimated that a higher percentage of their patients with diabetes received a serum creatinine test than did general practitioners. Conversely, general practitioners estimated that a higher percentage of their patients with diabetes received a dilated retinal exam.

o Respondents in solo or group private practices estimated that higher percentages of their patients with diabetes received blood tests (including serum creatinine, hemoglobin A1c and lipid profiles) than did those respondents in hospital, HMO or community based practices.

o Respondents with a greater number of active patients with diabetes in their practice said they administer several procedures to more of these patients than did those with fewer patients with diabetes. These procedures included blood tests (serum creatinine and lipid profiles) and urine

11

Page 12: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

tests (i.e., spot urine albumin/creatinine ratio and the 24 hour urine for albumin or protein excretion).

Respondents were also asked to identify the procedures they administer to patients with hypertension and the percentage of these patients who receive each procedure (Question 5). The table below summarizes these responses. As they do for their patients with diabetes, nearly all respondents said they use serum creatinine and lipid profiles for patients with hypertension and that nearly all of these patients receive these annually. Also similar to diabetes treatment is the infrequent use of 24-hour urine tests; just one-quarter of patients with hypertension receive this procedure annually.

Tests Routinely Used for Patients with Hypertension(Percent)

ProcedurePCPs

that use itPatients that get

annuallySerum Creatinine 97 91Lipid Profile 98 91Dipstick for Proteinuria 85 77Dipstick for Microalbuminuria 49 55Foot Exams 72 51Dilated Retinal Exam 63 47Spot Urine Albumin/Creatinine Ratio 50 42Hemoglobin A1c 62 4024 hour Urine for Albumin or Protein 61 2224 hour Urine for Creatinine Clearance 64 22

o More primary care providers in the pilot sites said they use a 24 hour urine for creatinine clearance and the spot urine albumin/creatinine ratio than did those in the control sites.

o Fewer non-physicians than physicians said they use serum creatinine, lipid profiles, dipstick for proteinuria and dilated retinal exam. In addition, physicians said they use serum creatinine with a greater percentage of their patients than did non-physicians.

o Fewer internists than general practitioners said they use a dipstick for microalbuminuria for their patients with hypertension. Internists estimated they administer foot exams to a greater percentage of their diabetes patients than did general practitioners.

o More primary care providers with higher than average numbers of patients with diabetes or hypertension said they use dilated retinal exams, spot urine albumin/creatinine ratio and serum creatinine than practitioners with fewer of these patients. These providers also said they used these tests with a greater percentage of their patients with diabetes than did physicians with fewer patients with diabetes.

Respondents were asked to indicate the effectiveness of early detection and treatment in stopping or slowing the progression of CKD in patients with hypertension, diabetes or polycystic kidney disease using a four-point scale ranked from “not at all effective” to “very effective” (Question 8). A majority of respondents said such treatment would be “very effective” in patients with diabetes (66%) and hypertension (64%). Only 18% of respondents felt such treatment was “very effective” for patients with polycystic kidney disease.

o More non-physicians than physicians said early detection of kidney disease would be “very effective” with patients with diabetes and polycystic kidney disease.

12

Page 13: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

When asked to identify the guidelines that have influenced their CKD treatment practices and procedures (Question 11), most respondents named the American Diabetes Association (ADA) Position Statement on Nephropathy (81%) and the Joint National Committee Report VI (JNC VI) (78%). Fewer respondents indicated being influenced by other guidelines. Just 21% of respondents identified the Kidney Disease Outcomes Quality Initiative (K/DOQI) Risk Stratification, 12% the Veterans Administration (VA) Guidelines, and 5% PARADE.

o More non-physicians, internists and those in HMO, community or hospital-based practices said the ADA Position Statement on Nephropathy was influential than did physicians, general practitioners and those in private practice.

o More physicians than non-physicians said the JNC VI had an influence on their treatment practices and procedures.

o Although mentioned by fewer respondents than other guidelines, more internists than general practitioners said K/DOQI was an influential guideline.

13

Page 14: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

Communicating about CKD and its Risk Factors

Respondents were asked to indicate which of several negative outcomes of poorly controlled diabetes they routinely emphasized in their conversations with patients who have diabetes (Question 3). Nearly all respondents said they discuss heart attack (98%) and kidney disease/kidney failure (98%), and the vast majority said they also discuss blindness (92%), stroke (89%), premature death (82%) and amputation (82%). About a third of respondents said they discuss other complications such as neuropathy/peripheral neuropathy (11%), sexual dysfunction/impotence/erectile dysfunction (8%) and peripheral vascular disease (3%).

o More pilot site respondents said they discuss kidney disease (99%) with their patients than did control site respondents (96%). However, more control site respondents said they discuss amputation (87%) than did pilot site respondents (80%).

o Among physicians, every internist said they discuss heart attack as a complication compared to nearly all (98%) general practitioners. Overall, more physicians (99%) than non-physicians (95%) said they discuss heart attack as a complication of poorly controlled diabetes; more physicians (84%) than non-physicians (77%) also said they discuss premature death.

o Practice setting affected the number of respondents who said they discuss stroke and premature death with their patients with diabetes. More respondents in group or private practice said they discuss these complications than did respondents in other practice settings.

Respondents also were asked to indicate which of several negative outcomes of poorly controlled hypertension they routinely emphasized in their conversations with patients who have hypertension (Question 6). Nearly all said they discuss stroke (99%) and heart attack (98%) and the vast majority said they also discuss kidney disease/kidney failure (94%), premature death (88%), and congestive heart failure (86%). About a third said they discuss blindness (32%) and less than a quarter discuss amputation (17%) or other complications (13%). Erectile dysfunction/impotence was the most common other complication respondents said they discuss.

o More pilot site respondents said they discuss heart attack (99%) with their patients than did control site respondents (96%).

o Among physicians, more internists said they discuss kidney disease (97%) as a complication compared to general practitioners (90%). More internists (92%) also said they discuss congestive heart failure than did general practitioners (85%). Overall, more physicians than non-physicians said they discuss stroke, heart attack and congestive heart failure than do non-physicians.

o More respondents in group or private practice said they discuss heart attack or congestive heart failure as complications than did respondents in other practice settings.

When asked to indicate how often they discuss kidney function tests with their at-risk patients (Question 13), one-half of respondents (51%) indicated they always discuss test results with patients, regardless of the outcome. Thirteen percent said they always discuss results when the test reveals a problem and one-third said they sometimes discuss the results (34%). Just two percent said they never discuss kidney function test results with patients and virtually all of these respondents were non-physicians.

14

Page 15: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

o More pilot site than control site respondents and those respondents with fewer patients with diabetes and/or hypertension said they always discuss test results when a problem exists.

Respondents were asked to indicate how often they discuss CKD with at-risk patients using a five-point scale of “Always—during every visit,” “Frequently—at nearly every visit,” “Occasionally-during some visits,” “Rarely,” and “Never” (Question 14). Just 4% said they never or rarely discuss CKD with at-risk patients. In all, 16% said they always discuss CKD with at-risk patients, 40% said they frequently do and 40% said they occasionally do.

o Internists were more likely than general practitioners to discuss CKD with their at-risk patients during every office visit.

o Although few respondents gave these responses overall, more nurse practitioners and physician assistants said they rarely (5%) or never (2%) discuss CKD with at-risk patients.

o Respondents with a higher than average number of African American patients were more likely than those with fewer African American patients to mention CKD at nearly every visit rather than during some visits (45% compared to 36%).

When asked to indicate which CKD topics they routinely discuss with at-risk patients, virtually all respondents said they discuss the importance of controlling hypertension and diabetes to prevent progression of kidney disease (Question 15). As shown in the table below, most respondents also discuss the severity and complications of kidney disease. About two-thirds of respondents said they discuss kidney function, the effect of CKD on organs and other prevention methods.

Topics Discussed with Patients At Risk for CKD(Percent)

The importance of controlling hypertension and diabetes to prevent progression of kidney disease

97

The severity and/or possible complications of kidney disease 87How kidneys function and why they are important 66Other ways of preventing or slowing its progression 64How kidney disease affects various organs and organ systems 61Options for treatment of kidney failure 56

o More internists than general practitioners said they discuss treatment options in the event of kidney failure, whereas more non-physicians than physicians said they discuss the impact CKD has on other organs.

o More of the solo or group private practice respondents stressed the importance of controlling hypertension and diabetes and other ways of preventing or slowing the progression of CKD than those representing the HMO, community or hospital-based practices.

Respondents were asked to rate their confidence in their ability to explain CKD to patients in an understandable manner using a 10-point scale where 10 means they felt very confident and a 1 means they did not feel confident at all (Question 16). Respondents indicated they were fairly confident, with a mean score of 7.37 on the 10-point scale. However there is room for improvement, as only one-quarter of respondents rated their confidence very highly (25% gave a 9 or 10).

15

Page 16: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

o In general, as specialized medical training increased so did ratings of confidence.. Non-physicians had a mean confidence rating of 6.7 whereas physicians had a 7.8. Similarly, general practitioners’ mean confidence rating was 7.4 and internists’ was 7.8.

o Respondents from group or solo private practices rated their confidence higher than those from HMO, community or hospital-based practices.

o Respondents who have a higher number of patients with diabetes and/or hypertension rated their confidence higher than those with fewer patients.

When asked if they believed they had sufficient tools and materials to discuss kidney disease with patients (Question 17), only one-third said yes (33%). Approximately two-thirds said they did not (44%) or were not sure if they did (22%).

o More internists believed they have enough tools and materials than did general practitioners. Similarly, more physicians believed they have sufficient tools and materials than did nurse practitioners and physician assistants.

o Compared to those with fewer patients, more respondents with a higher number of patients with hypertension and/or diabetes believed they had sufficient materials and tools.

o More respondents representing HMO, community or hospital-based practices believed they did not have sufficient materials than those in private practice settings.

From a list of materials, respondents were asked to select the tools they use while explaining CKD to patients (Question 18). As shown in the table below, respondents most often indicated they use brochures (47%), followed by posters (27%) flowcharts (23%), models (19%) and Internet sites (17%). Interestingly, one-quarter of PCPs (26%) said they do not use any materials whatsoever in their patient discussions.

Tools and Materials Used to Discuss CKD with Patients(Percent)

Brochures 47Posters 27Flowsheets or flowcharts 23Models 19Internet sites 17Videos 5Drawings/Illustrations 4Handouts 2Verbal Instruction 2Journal Articles 1Some other material 4Not applicable/I do not use any material when discussing CKD with my patients

26

o More pilot site (28%) than control site respondents (20%) said they do not use any materials when discussing CKD with their patients.

o More physicians said they used models and/or drawings and illustrations than did non-physicians. In contrast, physician assistants and nurse practitioners more often said they used Internet sites than did physicians.

16

Page 17: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

o More respondents from group or solo private practices reported that they use posters and models, while more of those in HMO, community or hospital-based practices said they use the Internet and video tools.

o More respondents with higher numbers of patients with diabetes and/or hypertension said they used posters, flowcharts and models than those with fewer patients with either condition.

Respondents were asked to indicate whether the discussion of CKD in the medical community in the last year had increased, decreased or stayed about the same (Question 19). Nearly one-half of respondents (45%) said it had stayed the same, about one-third (38%) said it had increased and 2% said it had decreased. One in six respondents (15%) said they did not know or were not sure.

o More PCPs in the Baltimore pilot site said there had been an increase in the discussion of CKD in the medical community in the past year whereas more respondents in Atlanta and Cleveland said there had been no change.

o Physicians, and especially internists, were more likely to cite increases in industry discussion of CKD

Respondents were asked to indicate whether their patients’ awareness of CKD had increased, decreased or stayed about the same over the past year (Question 20). A majority (57%) indicated their patients’ awareness had stayed about the same over the past year, 29% said it has increased and 1% said it had decreased. More than one in 10 respondents (13%) said they did not know or were not sure.

Respondents were asked to indicate whether they had heard more, less or about the same amount about CKD in the past year (Question 21). One-half said they had heard about the same amount and 43% said they had heard more.

o Internists were more likely to indicate they had heard more about CKD recently than in previous years. A few more general practitioners were unable to say (5% compared to 1% of internists).

17

Page 18: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

Appendix: Survey Instrument

18

Page 19: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

Please fax to: _________________________ @ (_____) _____ - __________ ID # ______________ (508 – 517)

Dear Primary Care Provider:

Thank you for assisting the National Institutes of Health of the Department of Health and Human Services in this research study. The information you provide will be kept confidential, and will not be disclosed to anyone but the researchers conducting this study, except as otherwise required by law. All identifying marks will be removed from your completed survey to ensure the anonymity of your response. If you have any questions regarding this survey, please don’t hesitate to email me at [email protected]. As a thank you to all respondents, we will be pleased to provide a summary of the final results when they are available.

Many of the questions below ask about current procedures in your practice – please answer these questions based on what your practice currently does and not what you would like to do for your patients. Your completed form should be faxed to: XXX XXX XXXX.

1. Approximately how many active diabetes patients are there in your practice, i.e. patients for whom you personally provide regular care and who you see at least once per year? (Please give us a rough estimate.)

_____________ [WRITE IN NUMBER](108 – 111)

2. Please estimate what percentage of these active diabetes patients receives each of the following procedures at least once a year as part of their routine care for diabetes. (RECORD A PERCENTAGE AT THE RIGHT OF EACH PROCEDURE.)

Percentage I do not use this testa) Dilated Retinal exam ............................................... ______% (122-24).................b) 24 hour urine for creatinine clearance...................... ______%............................c) 24 hour urine for albumin or protein excretion........ ______%............................d) Spot urine albumin/creatinine ratio.......................... ______%............................e) Dipstick for proteinuria/urinalysis........................... ______%............................f) Dipstick for microalbuminuria................................. ______%............................g) Lipid profile.............................................................. ______%............................h) Hemoglobin A1c...................................................... ______%............................i) Serum Creatinine ..................................................... ______%............................k) Foot exams............................................................... ______%............................l) Other (Specify): __________________................... ______% (152-54) .................

3. When discussing negative outcomes of poorly controlled diabetes in your conversations with patients who have diabetes, which of the following do you routinely emphasize? (CHECK ONE BOX AT THE RIGHT OF EACH OUTCOME.)

Yes No Yes NoStroke?..........................................1 .........2 Blindness? .......................................1 .........2

Amputation? .................................1 .........2 Kidney disease/kidney failure? .......1 .........2

Premature death? ..........................1 .........2 Some other outcome? ......................1 .........2

Heart attack? ................................1 .........2 (Specify):_________________

Not applicable/I do not counsel patients regarding their diabetes. ............ (168)

19

Sincerely,

Thomas Hostetter, MDDirector, National Kidney Disease Education Program

Page 20: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

20

Page 21: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

4. Approximately how many active hypertension patients are there in your practice? i.e. patients for whom you personally provide regular care and who you see at least once per year? (Please give us a rough estimate.)

_____________ [WRITE IN NUMBER](208 – 211)

5. Please estimate what percentage of these active hypertension patients receives each of the following procedures at least once a year. (RECORD A PERCENTAGE AT THE RIGHT OF EACH PROCEDURE.)

Percentage I do not use this testa) Dilated Retinal exam ......................................................______% (222-24)....................b) 24 hour urine for creatinine clearance.............................______%..............................c) 24 hour urine for albumin or protein excretion...............______%..............................d) Spot urine albumin/creatinine ratio.................................______%..............................e) Dipstick for proteinuria/urinalysis...................................______%..............................f) Dipstick for microalbuminuria........................................______%..............................g) Lipid profile.....................................................................______%..............................h) Hemoglobin A1c.............................................................______%..............................i) Serum Creatinine ............................................................______%..............................k) Foot exams......................................................................______%..............................l) Other (Specify): __________________..........................______% (252-54) ...................

6. When discussing negative outcomes of poorly controlled hypertension in your conversations with patients who have hypertension, which of the following do you routinely emphasize? (CHECK ONE BOX AT THE RIGHT OF EACH OUTCOME.)

Yes No Yes NoStroke?..........................................1 .........2 Congestive heart failure? .................1 .........2

Amputation? .................................1 .........2 Blindness? .......................................1 .........2

Premature death? ..........................1 .........2 Kidney disease/kidney failure? .......1 .........2

Heart attack? ................................1 .........2 Some other outcome? ......................1 .........2

(Specify):_________________

Not applicable/I do not counsel patients regarding their hypertension. ........

7. For each risk factor below, please check the box that best indicates how much you believe the risk factor increases a patient’s risk for CKD, i.e., chronic kidney disease or chronic renal insufficiency. (CHECK ONE BOX AT THE RIGHT OF EACH RISK FACTOR.)

Does not increase risk at all

Increases risk

slightly

Increases risk

moderately

Increases risk

greatlya) Race - African American..................................1 ..............2 ................3 ..............4 (308)

b) Gender – Male...................................................1 ..............2 ................3 ..............4

c) Diabetes.............................................................1 ..............2 ................3 ..............4

d) Hypertension.....................................................1 ..............2 ................3 ..............4

e) Urinary tract infections.....................................1 ..............2 ................3 ..............4

f) Kidney stones....................................................1 ..............2 ................3 ..............4

g) Smoking............................................................1 ..............2 ................3 ..............4

21

Page 22: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

h) Family history of kidney failure........................1 ..............2 ................3 ..............4

i) Multiple pregnancies.........................................1 ..............2 ................3 ..............4 (316)

22

Page 23: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

8. Based on your experience and knowledge, how effective is early detection and treatment in stopping or significantly slowing the progression of CKD in each of the following conditions? (CHECK ONE BOX AT THE RIGHT OF EACH CONDITION.)

Early detection and treatment is…Not at all effective

Slightly effective

Moderately effective

Very effective

a) CKD in Diabetes patients.............................1 ................2 .................3 ...............4 (317)

b) CKD in Hypertension patients.....................1 ................2 .................3 ...............4

c) Polycystic kidney disease.............................1 ................2 .................3 ...............4

9. Consider the following example: You have a 65-year-old Caucasian woman patient who weighs 50 kg/110 lbs and has had diabetes (HbA1C 8.0) and hypertension (BP 138/90) for several years. For each index below, please indicate the level at which you would make the diagnosis of CKD.

a) Creatinine level............................_____ . _________ (mg/dl) Do not use this index

b) Albumin/creatinine ratio............._______________ (mg/mg) Do not use this index

c) Proteinuria..................................._______________ (mg/24 hrs) Do not use this index

d) GFR............................................._______________ (ml/min) Do not use this index

10. Based on the diagnosis of early CKD in the patient above, which of the following steps would you typically do? (CHECK ONE BOX AT THE RIGHT OF EACH STEP.)

Yes No Yes NoAim for lower blood pressure?. (341) ....1 .........2 Caution patient against NSAID use?.........1 ..... .2

Measure lipid profile?.............................1 .........2 Counsel patient to reduce protein intake? .1 ..... .2

Consider use of diuretics?.......................1 .........2 Provide educational materials?..................1 ..... .2

Aim for tighter glycemic control?...........1 .........2 Refer the patient to a specialist?................1 ..... .2

Prescribe ACE inhibitors or ARBs ....1 .........2 Monitor kidney function more closely?. . ..1 ..... .2

(Angiotensin Receptor Blockers)?....1 .........2 Take some other step?...............................1 ..... .2

Prescribe Calcium Channel Blockers?.............................................1 .........2

(Specify): _______________________

11. Of the following guidelines, which have influenced your practices and procedures for treatment of CKD? (CHECK ONE BOX AT THE RIGHT OF EACH GUIDELINE.)

Yes No Yes NoAmerican Diabetes Association

(ADA) Position Statement on Nephropathy?................................1 ......2

PARADE? .................................................1 .....2

Kidney Disease Outcomes Quality Initiative (K/DOQI) Risk Stratification? 1 .....2

Joint National Committee Report VI (JNC VI)? ....................1 ......2

Veterans Administration (VA) Guidelines? .............................................1 .....2

12. Do you estimate GFR from serum creatinine using a calculation? (CHECK ONE BOX.)

4 Always 2 Sometimes 0 Not applicable/My lab reports GFR to me3 Often 1 Never

13. How often do you discuss results of tests of kidney function with patients who are at risk for CKD? (CHECK ONE BOX.)

4 Always, regardless of result 2 Sometimes3 Always, when there’s a problem 1 Never

14. How often do you typically discuss CKD with at-risk patients? (CHECK ONE BOX.)

5 Always—during every visit 2 Rarely (365)

23

Page 24: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

4 Frequently—at nearly every visit 1 Never3 Occasionally—during some visits

24

Page 25: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

15. Of the following topics, which do you routinely discuss with patients at risk for CKD? (CHECK ONE BOX AT THE RIGHT OF EACH TOPIC.)

Yes NoHow kidneys function and why they are important?................................................1 ...........2 (408)

How kidney disease affects various organs and organ systems?.............................1 ...........2

The severity and/or possible complications of kidney disease?...............................1 ...........2

Options for treatment of kidney failure?..................................................................1 ...........2

The importance of controlling hypertension and diabetes to prevent progression of kidney disease?.........................................................................1 ...........2

Other ways of preventing or slowing its progression?.............................................1 ...........2

Not applicable/I do not usually discuss CKD with at-risk patients................................

16. On the following scale from 1 to 10, please mark the number that best indicates how confident you are that you can explain CKD to your patients in a way that they understand. (CHECK ONE BOX.)

Do Not Feel AtAll Confident

Feel Very Confident

1 2 3 4 5 6 7 8 9 10

17. Do you have sufficient tools and materials to discuss kidney disease with your patients? (CHECK ONE)

1 Yes 2 No3 Not sure

18. Of the following materials, which do you typically use to discuss CKD with your patients? (CHECK ONE BOX AT THE RIGHT OF EACH MATERIAL.)

Yes No Yes NoBrochures?....................................1 .........2 Internet sites? ..................................1 .........2

Posters? ........................................1 .........2 Videos? ............................................1 .........2

Flowsheets or flowcharts? ............1 .........2 Some other material? .......................1 .........2

Models? ........................................1 .........2 (Specify):_________________

Not applicable/I do not use any material when discussing CKD with my patients........................

19. In the last year, has the discussion of CKD in the medical community increased or decreased? (CHECK ONE BOX.)

1 Increased2 Decreased3 Stayed about the same4 Don’t know/not sure

20. In the last year, has your patients’ awareness of CKD increased or decreased? (CHECK ONE BOX.)

1 Increased2 Decreased3 Stayed about the same4 Don’t know/not sure

21. In the last year, have you heard more or less about CKD than you did in previous years? (CHECK ONE)

25

Page 26: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

1 Have heard more (434)

2 Have heard less3 About the same4 Don’t know/Not sure

26

Page 27: Respondent Characteristics

DRAFT REPORT. DO NOT CITE OR QUOTE.

Please tell us a little about yourself.

22. Are you a… (CHECK ONE BOX.)

1 Family physician (435)

2 General Practitioner3 Internal medicine (What is your sub-specialty, if any? )

4 Nurse practitioner Skip to Question 245 Physician’s assistant Skip to Question 24

23. Are you board certified? (CHECK ONE BOX.) 1 Yes 2 No

24. How long have you been in practice since completing your residency/training? (CHECK ONE BOX.)

1 Less than 5 years2 6-10 years

3 11-15 years4 More than 15 years

25. What is your practice setting? (CHECK ONE BOX.)

1 Solo or group private practice2 Community health clinic3 Hospital based

4 HMO/managed care5 Other (Specify): ___________________

26. Please tell us about the composition of your patient population by assigning rough percentages to the following categories. [The total should equal 100%.]

African American/Black..................______%Asian or Pacific Islander.................______%Caucasian/White..............................______%Hispanic...........................................______%Native American..............................______%Other................................................______%

TOTAL 100%

27. What is the zip code of your practice? _______________________

Thank you for your participation. To receive your honorarium please complete the information below and fax this completed form to XXX-XXX-XXXX.

Please send honorarium to: (PLEASE PRINT)

Name:

Address:

City: State: Zip Code:

Would you like NIH to send you a summary of these results when available? 1 Yes 2 No (476)

Public reporting burden for this collection of information is estimated to average 10 minutes per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. An agency may not conduct or sponsor, and a person is not required to respond to, a collection of information unless it displays a currently valid OMB control number. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to: NIH, Project Clearance Branch, 6705 Rockledge Drive, MSC 7974, Bethesda, MD 20892-7974, ATTN: PRA 0925:0515. Do not return the completed form to this address.

27