Skip to main content
. 2015 Dec 22;5:150–161. doi: 10.15256/joc.2015.5.61

Table 1.

Key methodologic elements of research protocol.

Major category Methodologic considerations Details Methodologic elements
1. Design Research design Retrospective or historic cohort study
2. Population and sampling Location Ten practice-based research networks across Canada; located in eight Canadian provinces: British Columbia, Alberta (two networks), Manitoba, Ontario (three networks), Quebec, Nova Scotia and Newfoundland and Labrador
Sampling Sampling method Recruitment of PHC networks, sites, and providers; collection of patient data from the participating PHC providers
PHC setting Sampling frame Consenting PHC networks, sites, and providers
Sampling method Non-random recruitment; based on interest in participating in CPCSSN dataset
Sampling size 82 PHC sites from 10 PHC networks
PHC providers Sampling frame 39,392 family physicians [28] and 3,655 nurse practitioners in Canada in 2013 [29]
Approximate total of 43,047 PHC providers in Canada in 2013
Sampling method Non-random recruitment; based on interest in participating in CPCSSN dataset
Sampling size 477 PHC providers recruited and contributing data
PHC patients Sampling frame 35,345,000 population (and therefore potential PHC patients) in Canada in 2013 [30]
Sampling method Non-random collection of data; consistent and consecutive data collection
Sampling size 600,565 PHC patients included and contributing data
Rationale for sample size Utilized all eligible, adult patients included in the CPCSSN database
3. Data and definition Data collection Source of data Entire electronic medical record; provider documentation at each encounter
Method of data collection Extraction of de-identified and anonymized EMR data
Coding Morbidity coding ICD-9
Time Time period of data source Length of EMR and ending on September 30, 2013 (date of data extract)
Length of recruitment period for patients for each PHC provider Recruitment began in 2008 and ended as of September 30, 2013 (date of data extract); length of retrospective data entered into the EMR system for each patient varied by PHC providers
Dates of data collection All data were extracted as of September 30, 2013
Morbidity time focus Ever diagnosed in the EMR (lifetime morbidity)
Definitions Definition of multimorbidity Two or more chronic disease diagnoses; creation of mutually exclusive categories of multimorbidity (patients with 2, 3, 4 and 5+ chronic disease diagnoses)
Definition of chronic diseases Chronic diseases were identified according to WHO definition of chronic and based on burden among populations; list of 20 chronic disease categories and corresponding ICD-9 codes are included in Table 2
Operational definition of the count of chronic diseases PHC providers recorded diagnostic code during clinical encounter (not for research purposes); double counting was avoided by identifying the first chronic disease diagnoses of those chronic diseases on list
4. Outcomes Results Outcomes reported Prevalence of single morbidity; prevalence of multiple morbidities; prevalence estimates stratified by relevant covariates; commonly occurring clusters of multiple chronic diseases; time-until-event information on both onset and advancing multimorbidity
Confounders controlled Site characteristics (geographical location); provider characteristics (age, sex); patient characteristics (age, sex, residential geographic location)
Results presented Prevalence of multimorbidity; characteristics of adult patients living with multimorbidity; prognosis of multimorbidity over time

CPCSSN, Canadian Primary Care Sentinel Surveillance Network; EMR, electronic medical record; ICD-9, International Classification of Disease, 9th Revision; PHC, primary healthcare; WHO, World Health Organization.