Sage Journals: Discover world-class research

Abstract

US health systems are required to conduct community health needs assessments (CHNAs). These assessments often rely solely on publicly reported disease prevalence data. The objective of this study was to identify a feasible qualitative method that can be implemented into CHNAs to confirm existing information and enhance them with new data and community narrative. Freelisting interviews were conducted in 2017 throughout a county in Pennsylvania. Individuals listed (1) things that improve health, (2) health problems, and (3) health barriers. Responses were grouped into like terms, Smith’s salience index was calculated for each, subgroup variation was assessed. Community-reported health priorities were identified by salient terms juxtaposed with publicly reported disease prevalence data. 98 respondents throughout the county participated in freelisting interviews. Demographics resembled the population: median age 55, 46% female, 88% Caucasian, 9% uninsured. Salient terms that improve health included: “physical activity”, “healthy eating”, and “health services”. Salient terms describing health problems included: “cardiovascular disease” and “spine and joint disease”. Centers for Disease Control and Prevention (CDC)-reported chronic disease prevalence in the county included heart disease mortality (22% of deaths) and arthritis (26% of adults). Salient barriers to health included: “weight, physical activity, and dietary concerns”, as well as “occupational concerns”. This is in line with 2017 County Health Rankings related to obesity (reported 29%), however, while 10% were uninsured in the community, this was not a salient term. Freelisting can augment and inform CHNAs. In this single-county study, freelisting responses supported publicly available disease/mortality prevalence data and provided insight into community perceptions of health issues.

Keywords

community health needs assessment community health qualitative methods applied research program Enhancement

Background

Community health needs assessments (CHNA) are required and necessary to inform strategic planning of health services at non-profit hospitals and community health centers.¹ Since the Affordable Care Act mandated CHNAs in 2010, these systems have produced reports that vary in their ability to inform community health practice.² According to the Centers for Disease Control and Prevention (CDC) (2016) many CHNAs rely on primary and secondary data and may include data from surveys, focus groups, and interviews which shed light on the community’s perceptions of what impacts health and disease. However, these approaches have some limitations and can be very time consuming. Our study was conducted to help address this challenge by utilizing freelisting, an accessible and efficient data collection approach that allows researchers to quickly assess community perceptions and priorities.

Freelisting is an anthropological technique in which participants list items in response to questions.^3–8 Previous studies have utilized this method to prioritize service quality dimensions,⁹ identify barriers to self-efficacy,¹⁰ and allocate for mental health services,¹¹ in resource-limited settings. Community health centers with limited research capacity can feasibly obtain valuable qualitative data using freelisting, which can complement publicly-reported disease prevalence or mortality data and inform planning and the approach, unlike traditional interviews or surveys, allows researchers to appreciate salient issues for subgroups of people in a sample. Freelisting interviews also involve a lower response burden than most surveys because they can be conducted verbally, without requiring reading or writing, and they allow respondents to list things that come to mind without restriction. In communities with low literacy and low levels of education, this interview technique can be particularly valuable.^12,13

Like many community health centers in the United States, the investigators of this study faced challenges when assessing the needs of the community. Publicly-reported disease prevalence and mortality data were available. However, the health priorities of community residents were not well represented prior to this study. In order to be effective, plans for disease prevention and control need to be tailored to the priorities of individuals in a community.^14,15

Purpose

Therefore, the objective of this study was to examine whether freelisting can augment existing CHNA practices by more fully describing community needs, investigating community health beliefs, informing follow-up surveys or qualitative studies, and supporting local health planning.

Methods

Study Population & Settings

Adults were recruited to participate in freelisting interviews in rural and urban areas across a single county in Pennsylvania. Community residents were eligible for participation if they were at least 18 years of age or older and resided within the county. An intercept interview approach,⁷ was employed in which participants were approached and recruited on the spot from public places including shopping centers, town squares, public housing units, and health centers. Participants provided informed consent to be interviewed in accordance with a research protocol that was approved by the xxxx Institutional Review Board.

Study Design

A group of seven interviewers were trained by the research team to conduct freelisting interviews. The interviewers included local county residents, community volunteers, academic faculty and research staff who responded to an email call for volunteers or heard about the opportunity from a research team member. Structured interviews were conducted using the freelisting method.^3–8,16,17 Interviewers instructed participants to list all the words that came to mind in response to the prompt, stating as many words or phrases that they could offer in response, with no limit on the number or type of words. The interview prompt involved asking participants to do the following:

• List the things that improve your health.

• List the health problems that affect you.

• List the things that make it hard for you to stay healthy.

Following the freelisting prompts, respondents were asked to answer a set of demographic questions regarding zip code, year of birth, health insurance status, race, and gender (Supplement 1). Interviews lasted 5–10 minutes. The interviewer documented participant responses in the order in which they were stated, and research staff subsequently entered the responses into a database.

Analysis

Descriptive statistics were calculated to for the age, gender, race, insurance status, and geographic region of participants. These descriptive statistics were then compared to relevant publicly reported data (US Census Bureau Quick Facts; US Census Bureau SAHIE Program). Freelisting responses were analyzed through a process of grouping synonymous terms. The final set of grouped terms were mutually exclusive.

The identification of synonyms and grouping of terms was conducted by a single investigator, and, subsequently, at least two co-investigators adjudicated the grouping assignments. Disagreements were resolved by team discussion and consensus decisions.

After terms were finalized and groupings completed, the most salient terms were identified using Smith’s salience index.¹⁷ The salience index, or score, is a function of the frequency with which respondents list a term and the order in which it is listed, such that the salience index can be calculated as follows:

S j = (\frac{(L - R_{j} + 1) / L}{N})

Where the salience of a term j is a function ofR_j, its rank in the list,L, the length of each list, andN, the number of lists in the sample.

After a salience index was calculated for each term, the terms in each of the three lists were sorted from highest to lowest salience index then plotted on a graph with salience on the y-axis and terms on the x-axis (each represented by a point). Terms were determined to be salient if their scores fell above the “elbow” on the curve on, that is the point where data leveled off and the curve began to flatten, as previously described.^7,16,18,19 The number of salient terms varied for each list due to differences in the distribution of salience scores. Stratified analyses were then completed, using the same methods, for men versus women, and adults ages 18 to 64 versus adults ages 65 and older. These analyses were completed using Anthropac 1.0 Freelists and R version 3.4.2. Salient community-reported health information was then compared to publicly reported vital statistics on disease prevalence, mortality and health behaviors.^20–23

Results

A total of 98 participants across a county in Pennsylvania were interviewed. Approximately 125 people were approached and invited to participate (roughly a 78% response rate). The median age of participants was 55, women made up 46%, 12% were Black or of a non-White race, and 9% were uninsured. The demographics of respondents are shown in Table 1 next to publicly-reported descriptive statistics for the county.^23,24

Table 1.

Demographics.

Characteristics	Study N = 98	Study, %	County %
Age
35 and under	23	24	39%^a
36–50	22	23	17%^a
51–65	29	30	23%^a
Over 65	24	24	21%^a
Gender
Male	49	50	49%^a
Female	45	46	51%^a
Other or missing	4	4
Race
Black	6	6	4%^a
White	86	88	94%^a
Other or missing	6	6	2%^a
Health insurance status
Insured	89	91	92%^b
Uninsured or unknown	9	9	8%^b

^aAt the county-level, publicly reported data on age, gender, and race were obtained from the US Census Bureau Quick Facts 2016 estimates.

^bCounty-level data on insurance status were obtained from the 2015 US Census Bureau’s Small Area Health Insurance Estimates (SAHIE).

Responses to the prompt “List the things that improve your health” included 127 unique terms, which investigators grouped into 11 terms. In response to interviewers’ request to “List the health problems that affect you”, participants shared 78 unique terms that were grouped into 13 terms. Finally, responses to the request to “List the things that make it hard for you to stay healthy” included 149 unique terms, which were grouped into 11 terms. Supplement 2: Grouped Terms illustrates the identification of synonymous terms.

The salient terms describing factors that improve health included: “Exercise and physical activity,” “Eating healthy food,” “Healthcare and health services,” “Sleep and rest,” “Stress reduction and relaxation”. All terms about health improvement are listed in Table 2, with salient terms in bold.

Table 2.

Community-Reported Terms for Factors that Improve Health.

Factors that Improve Health (Salient Factors in Bold)	Salience Score (Smith’s S)
Exercise & physical activity	0.373
Eating healthy food	0.354
Healthcare & health services	0.13
Sleep & rest	0.098
Stress reduction & relaxation	0.088
Smoking cessation	0.055
Medications & medical devices	0.046
Stop drinking alcohol	0.033
Stop drug abuse	0.21

Salient terms describing health problems included: “Cardiovascular disease,” “Diabetes,” “Spine and joint disease,” “Weight, activity and dietary concerns,” “Cancer”. These terms describing health problems in the community are listed in Table 3 next to publicly reported vital statistics.^20,22

Table 3.

Community-Reported Health Problems.

Community-Reported Health Problems (Salient Factors in Bold)	Salience Score (Smith’s S)	Publicly-Reported Disease Prevalence & Mortality
Cardiovascular disease ^a	0.281	22% of deaths^a
Diabetes	0.143	13% of adults^c
Spine & joint disease	0.124	26% of adults have arthritis^c
Weight, activity, & dietary concerns	0.11	30% of adults are obese^c
Cancer ^a	0.08	21% of deaths^a
Asthma & lung disease^a	0.072	14% of adults^c
Smoking	0.068	17% of adults^c
Mental health condition	0.058	—
Allergies	0.051	—
Anxiety	0.049	—
Stress	0.048	—
Drug problem	0.035	2% of deaths^a
Alcohol problem	0.034	20% of adults^c

^aDenotes community reported health problems aligning with one of the top five causes of death in Cambia County in 2017–2018, which include, (1) #Diseases of heart, (2) Malignant Neoplasms, (3) Accidents (unintentional injury), (4) Cerebrovascular Disease, (5) Chronic Lower Respiratory Diseases.

^bMortality data were obtained from the Pennsylvania Department of Health 2015 Vital Statistics.

^cDisease prevalence and mortality data were obtained from the²⁰ County Data Indicators (CDC). A dash indicates that publicly reported data were unavailable at the county-level.

The salient terms encompassing barriers to health included: “Weight, activity, and dietary concerns,” “Occupational concerns,” “Time constraints,” “Stress and fatigue”, “Lack of Motivation”. These terms describing barriers to health are included in Table 4, alongside publicly reported county-level estimates on social determinants of health.^21,23

Table 4.

Community-Reported Terms for Barriers to Health.

Barriers to Health (Salient Factors in Bold Listed in Order – Highest First)	Salience Scores (Smith’s S)	Publicly-Reported Health Behaviors
Weight, activity, & dietary concerns	0.328	41% have inadequate access to physical activity opportunities*
Weight, activity, & dietary concerns	0.328	14% live with food insecurity*
Occupational concerns	0.184	27% have a long commute to work*
Time constraints	0.136	20% have no leisure time physical activity*
Stress & fatigue	0.094	—
Lack of motivation	0.076	—
Financial constraints	0.071	16% living in poverty**
(Barriers to) healthcare & health services	0.063	—
Smoking cessation	0.061	—
Aging	0.046	21% of the population is over age 65
Alcohol problems	0.026	42% of driving deaths involve alcohol
Lack of social support	0.023	—

Sources: *County Health Rankings and Roadmaps 2017, ** US Census Bureau Quick facts 2016 estimates. A dash indicates that publicly reported data were unavailable at the county-level.

Stratified analyses comparing responses of women versus men revealed key differences (Figure 1). While both women and men considered “physical activity” and “healthy eating” to be important for improving health, “healthcare and health services” were salient for women while “sleep and rest” were salient for men. Salient health problems among women and men included “cardiovascular disease”, “diabetes”, and “spine and joint disease”. “Weight, activity, and dietary concerns” also comprised a salient group of terms among both genders. “Cancer” was a salient health problem reported among women, whereas “asthma and lung disease” was a salient health problem among men. Among both women and men, salient barriers to health included “weight, activity, and dietary concerns”, as well as “occupational concerns”, and “stress and fatigue”. For women, salient barriers to health also included “financial constraints” and “time constraints”, whereas additional barriers to health that were salient among men involved limited “access to health services” and “lack of motivation”.

Figure 1.

Salient health priorities among women compared to men. Legend: Venn diagrams constructed using Smith’s salience indexes to determine similarities and differences regarding health priorities for men and women.

Analyses comparing responses of adults age 18 to 64 next to those of adults age 65 and older revealed similarities and differences (Figure 2). For all adults, “physical activity”, “eating healthy food”, and “healthcare and health services” were salient factors thought to improve health. But, for adults under 65, “sleep and rest” were also found to be salient for improving health, while “medications and medical devices” were salient for adults over 65. Salient health problems reported among all adults included “cardiovascular disease”, “diabetes”, as well as “spine and joint disease”. For adults under age 65, “smoking”, “cancer”, as well as “weight, activity, and dietary concerns” were also salient. Among adults ages 65 and older, additional salient health problems included “stress and fatigue” as well as “asthma and lung disease”. Salient barriers to health for all adults included “weight, activity, and dietary concerns”, as well as “stress and fatigue”. Among adults under age 65, additional salient barriers to health included “occupational concerns” and “time constraints”. For adults over age 65, “aging” was perceived as an additional salient barrier to health.

Figure 2.

Salient health priorities among adults ages 18 to 64 compared to adults age 65 and older. Legend: Venn diagrams constructed using Smith’s salience indexes to determine similarities and differences regarding health priorities for younger and older adults.

Discussion

This study demonstrated that the freelisting method can effectively contribute information to CHNAs that is not readily available in current public prevalence data, such as community perceptions of things that improve health. The study population also reported health concerns and perceptions that were congruent with publicly reported disease prevalence and mortality data with no notable differences. This demonstrates that inclusion of both methods in a CHNA verifies the health needs of the community while contributing unique information and narrative.

Recent attention to diseases of despair, including suicide, drug overdoses, and alcoholic liver disease,^25,26 did not appear to distort individual health perceptions and priorities in this study. Although the rates of such diseases are growing in the study population, they are not among the top 10 causes of death in the county.²² Respondents mentioned health concerns including “alcohol problem” and “drug problem”, but these were not among the salient terms describing community health priorities. Cardiovascular disease was among the salient health problems reported in this study and it is the leading cause of death in the county.²² The congruence between publicly reported disease prevalence and mortality and community-reported health priorities supported the validity of our study results.

Qualitative data from freelisting responses also complemented disease prevalence and mortality data by providing descriptions of community perceptions of disease, including what individuals feel they can do about disease, and what factors make it difficult for them to address their health concerns. For example, respondents who reported having a “bad back” or “bad knees” provided more detail to describe publicly reported rates of arthritis in adults,²⁰ and could inform future approaches to physical activity interventions.

Use of the freelisting method also offered several advantages over other methods of collecting qualitative data. Whereas surveys, qualitative interviews, and focus groups often require highly skilled interviewers or facilitators,^7,27 our freelisting interviewers required a very brief training and only needed the capacity to effectively communicate a simple prompt and record responses. Data collection and analysis were efficient in this study. Listing of responses is less labor intensive than transcribing complex interviews or focus group discussions. Analysis involved grouping and scoring of terms that were pertinent to the research question,⁷ limiting extraneous data, without requiring exploration of transcripts from interviews or focus groups.²⁸ Lastly, because the freelisting methods we used in this study involved a low response burden for interviewees, non-response bias that is often seen with other qualitative methods,^29,30 was largely avoided.

Limitations

Despite the advantages of the freelisting method, this study had some limitations. Given the small sample size of 98, we cannot claim that interview results are completely representative of all 140,000 residents in the county. Additionally, while the vast majority of people approached to be interviewed agreed to participate (approximately 78%), we do not know how many people saw the interviewers, and perhaps unsure of what they were doing, avoided the encounter by crossing the street or changing directions. Thus both, response bias and social acceptability bias could have played a role in results. However, the congruence between community-reported health priorities and publicly reported disease prevalence and mortality data make a case that results could be a fair representation of the county. Also, we were unable to make comparisons between all subgroups as we did not reach saturation in every group (generally you need at least 20 respondents per subgroup), for example while 89 respondents identified as White only six identified as Black. Additionally, qualitative methods such as freelisting require interpretation of responses, which can lead to bias. In this case, the investigators’ interpretation was influenced by the community health planning perspective, which involved grouping terms into disease areas that can be prevented or controlled through health interventions. Interpretation of responses could have resulted in different groupings if the study were completed by investigators in a different discipline. Nonetheless, the investigators described the methods of this study so as to make the analysis as transparent and replicable as possible. Other community health centers seeking to complement a needs assessment with qualitative data could benefit from using the freelisting method described in this study.

Implications for Research

Further work must include cost-benefit analysis to make sure that any CHNA additions are truly benefiting the community. Other directions may include modeling the steps necessary to include new methods into an existing CHNA structure. Freelisting and other approaches, like Group Concept Mapping³¹ or Intercept Interviews⁸ are efficient modes of data collection and should be consider as they may have utility in the context of CHNA.

This must include cost, expansion or addition of public health practitioner roles, and time considerations. Findings from this study may be used to inform future surveys or qualitative studies in this community. Lastly, if implemented, retrospective analyses should confirm the utility of any new methods and if they should be maintained.

Conclusions

This study demonstrated a reasonable means of obtaining useful qualitative data to inform community planning and population health practice. Freelisting methods were used to give community residents a voice in the assessment of health needs to allow them to participate in the prioritization process. Additionally, the freelisting method informed the development of an infographic (Supplement 3), which has been disseminated to the community. We hope this will inspire health priority conversations among community members and between community members and their providers.

Supplemental Material

Supplemental Material - Freelisting: A Technique for Enhancing the Community Health Needs Assessment

Supplemental Material for Freelisting: A Technique for Enhancing the Community Health Needs Assessment by Marquita Decker-Palmer, David Klodowski, Trina Thompson, Marianna Lanoue, Allison Messina, Diana Schroeder, Shiryl Barto, Billy Oglesby, and Rosemary Frasso in American Community Health Equity Research & Policy

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the 1889 Foundation, Inc. (No associated grant number).

Informed Consent

Participants provided informed consent to be interviewed in accordance with a research protocol that was approved by the Jefferson Institutional Review Board.

ORCID iDs

David Klodowski

Billy Oglesby

Rosemary Frasso

Supplemental Material

Supplemental material for this article is available online

Author Biographies

Marquita Decker-Palmer MD, PhD, MPH is the former Associate Director at the 1889 Jefferson Center for Population Health and is currently a health economist for Genentech.

David Klodowski MD, MPH is a former student at The Jefferson College of Population Health and Sidney Kimmel Medical College and is currently a resident physician at Beth Israel Deaconess Medical Center/Harvard Medical School.

Trina Thompson DrPH, MPH, BSN is the former Executive Director at the 1889 Jefferson Center for Population Health and is currently a healthcare contractor.

Marianna LaNoue PhD, MS is a former Associate Professor and Program Director at the PhD program in Population Health Science at Thomas Jefferson University and is currently a Professor of Statistics and Measurement at Vanderbilt University

Allison Messina MHPE is the Director of Evaluation at the Center for Health Promotion and Disease Prevention in Johnstown, PA, USA.

Diana Schroeder DNP, MSN, RN is the Director of Bullying Prevention Initiatives at the Center for Health Promotion and Disease Prevention in Johnstown, PA, USA.

Shiryl Barto MEd is the Manager at the Center for Health Promotion and Disease Prevention in Johnstown, PA, USA.

Billy Oglesby PhD, MBA, MSPH is an Associate Professor and interim Dean of the Jefferson College of Population Health.

Rosemary Frasso PhD, MSc, MSc is a Professor and the Director of the Public Health Program at the Jefferson College of Population Health.

References

Patient Protection and Affordable Care Act . 42 U.S.C. § 18001 et seq, 2010.

Pennel

McLeroy

Burdine

, et al. Nonprofit hospitals’ approach to community health needs assessment. Am J Public Health 2015; 105(3): e103–e113.

Borgatti

. Using Anthropac to investigate a cultural domain. Cult Anthropol Methods Newsl 1989; 1(2): 11.

Borgatti

. Cultural domain analysis. J Quantitative Anthropol 1994; 4(4): 261–278.

Borgatti

. ANTHROPAC 4.0. Natick, MA: Analytic Technologies, 1996.

Borgatti

. Elicitation techniques for cultural domain analysis. Enhanced Ethnographic Methods 1999; 3: 115–151.

Frasso

Keddem

Golinkoff

. qualitative methods: tools for engaging and empowering communities. Chapter 32 in Cnaan

Milofsky

(Eds). Handbook of community movements and local organizations. NewYork: Springer, 2018.

Keddem

Barg

Frasso

. Peer reviewed: practical guidance for studies using freelisting interviews. Prev Chronic Dis 2021; 18: E04.

Chowdhary

Prakash

. Prioritizing service quality dimensions. Managing Serv Qual Int J 2007; 17(5): 493–509.

10.

Bayliss

Steiner

Fernald

, et al. Descriptions of barriers to self-care by persons with comorbid chronic diseases. Ann Fam Med 2003; 1(1): 15–21.

11.

Bolton

Tang

. Using ethnographic methods in the selection of post-disaster, mental health interventions. Prehosp Disaster Med 2004; 19(1): 97–101.

12.

Ahmad

Barg

Bowles

, et al. Comparing perspectives of patients, caregivers, and clinicians on heart failure management. J Cardiac Failure 2016; 22(3): 210–217.

13.

Bennett

Switzer

Aguirre

, et al. ‘Breaking it down’: patient-clinician communication and prenatal care among African American women of low and higher literacy. Ann Fam Med 2006; 4(4): 334–340.

14.

Brownson

Baker

Boyd

, et al. A community-based approach to promoting walking in rural areas. Am J Preventive Med 2004; 27(1): 28–34.

15.

Hawkins

Catalano

Arthur

. Promoting science-based prevention in communities. Addict Behaviors 2002; 27(6): 951–976.

16.

Ryan

Nolan

Yoder

. Successive freelisting: Using multiple free lists to generate explanatory models. Field Methods 2000; 12(2): 83–107.

17.

Smith

. Using ANTHOPAC 3.5 and a spreadsheet to compute a free-list salience index. CAM 1993; 5(3): 1–3.

18.

Fiks

Gafen

Hughes

, et al. Using freelisting to understand shared decision making in ADHD: parents’ and pediatricians’ perspectives. Patient Educ Counseling 2011; 84(2): 236–244.

19.

Schrauf

Sanchez

. Using freelisting to identify, assess, and characterize age differences in shared cultural domains. J Gerontol Ser B: Psychol Sci Social Sci 2008; 63(6): S385–S393.

20.

Centers for Disease Control and Prevention . Diabetes – county data indicators. Available at: https://www.cdc.gov/diabetes/data/countydata/countydataindicators.html (2016, Accessed April 2, 2018).

21.

County Health Rankings & Roadmaps . Available at: www.countyhealthrankings.org (2017, Accessed April 2, 2018).

22.

Pennsylvania Department of Health . Vital Statistics: 2015 Death Statistics. 2015. Available at http://www.statistics.health.pa.gov/HealthStatistics/VitalStatistics/DeathStatistics/(2015, Accessed April 2, 2018).

23.

US Census Bureau. Quick facts . Available at https://www.census.gov/quickfacts (Accessed 2 April 2018).

24.

US Census Bureau. Small Area Health Insurance Estimates (SAHIE) Program . Available at https://www.census.gov/programs-surveys/sahie/data.html (Accessed April 2, 2018).

25.

Meit

Heffernan

Tanenbaum

, et al. Appalachian diseases of despair. Report for the appalachian regional commission. The Walsh Center for Rural Health Analysis National Opinion Research Center (NORC) at the University of Chicago, 2017.

26.

Stein

Gennuso

Ugboaja

, et al. The epidemic of despair among White Americans: trends in the leading causes of premature death, 1999–2015. Am J Public Health 2017; 107(10): 1541–1547.

27.

Tong

Sainsbury

Craig

. Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups. Int J Quality Health Care 2007; 19(6): 349–357.

28.

Desvousges

Smith

. Focus groups and risk communication: the “science” of listening to data. Risk Anal 1988; 8(4): 479–484.

29.

Asch

Jedrziewski

Christakis

. Response rates to mail surveys published in medical journals. J Clin Epidemiol 1997; 50(10): 1129–1136.

30.

Sheehan

. E-mail survey response rates: a review. J Computer-Mediated Communication 2001; 6(2): JCMC621.

31.

Rising

LaNoue

Gentsch

, et al. The power of the group: comparison of interviews and group concept mapping for identifying patient-important outcomes of care. BMC Med Res Methodol 2019; 19(1): 7. doi: 10.1186/s12874-018-0656-x.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.27 MB