Clustering and trajectories of key noncommunicable disease risk factors in Norway: the NCDNOR project

Noncommunicable diseases (NCDs) are a leading cause of premature death globally and have common preventable risk factors. In Norway, the NCDNOR-project aims at establishing new knowledge in the prevention of NCDs by combining information from national registries with data from population-based health studies. In the present study, we aimed to harmonize data on key NCD risk factors from the health studies, describe clustering of risk factors using intersection diagrams and latent class analysis, and identify long-term risk factor trajectories using latent class mixed models. The harmonized study sample consisted of 808,732 individuals (1,197,158 participations). Two-thirds were exposed to ≥ 1 NCD risk factor (daily smoking, physical inactivity, obesity, hypertension, hypercholesterolaemia or hypertriglyceridaemia). In individuals exposed to ≥ 2 risk factors (24%), we identified five distinct clusters, all characterized by fewer years of education and lower income compared to individuals exposed to < 2 risk factors. We identified distinct long-term trajectories of smoking intensity, leisure-time physical activity, body mass index, blood pressure, and blood lipids. Individuals in the trajectories tended to differ across sex, education, and body mass index. This provides important insights into the mechanisms by which NCD risk factors can occur and may help the development of interventions aimed at preventing NCDs.


Study Description The Norwegian Counties Study
The study included participants in three (out of 19 in total) Norwegian counties with a mainly rural settlement (Finnmark, Sogn og Fjordane, and Oppland).Originally, three rounds of data collection were conducted in the 1970s and 80s.These three first studies included anthropometric measurements, blood pressure measurements, non-fasting blood samples and a questionnaire.In Finnmark, the 1 st study was carried out in 1974-75, in Sogn og Fjordane in 1975-76, and in Oppland in 1976-78.For the first study, all residents aged 35-49 years were invited to participate.In addition, a 10% random sample of residents aged 20-34 was invited in Sogn og Fjordane and Oppland.In Finnmark, all residents aged 20-34 years were invited in four of the county's municipalities in addition to a 10% random sample in the remaining municipalities.The overall participation in the 1 st study was 88%. [1]he 2 nd study was carried out after three years in Finnmark (1978-79), and after five years in Sogn og Fjordane (1980-81) and Oppland (1981-83).The overall participation in the 2 nd study was 88%. [2]The following were invited to the 2 nd study: • o Everyone invited to the 10% random sample in the 1 st study still residing in Oppland aged 25-39 o A 5% random sample of residents aged 20-24 Minor modifications were made before the 3 rd study, which was carried out 1987-1988 in Finnmark, 1985-1986 in Sogn og Fjordane, and 1986-1988 in Oppland, but the main core of participants in the previous rounds were invited.This resulted in a sample of residents aged 20-49 years comparable to the samples aged 20-49 years in the 1 st and 2 nd study, and a prospective sample with two or three participations (aged from 20 years at the first participation to 63 years at third participation).For the cohort aged 45-49 years at the 1 st study, capacity constraints limited invitations to a 10% random sample. [3]The overall participation in the 3 rd study was 84%.In 2006-2008, a follow-up questionnaire was sent to all those still alive that had participated at least once previously and were born between 1925-1947 (Finnmark), 1926-1945 (Sogn og Fjordane) and 1927-1946 (Oppland) (i.e.aged 59-83 years.The overall participation in this 4 th study was 59%.The 4 th study was questionnaire based only.In NCDNOR, 94,022 participants of the Norwegian Counties Study were included (Figure 1).We excluded participants aged <20 years (n=454), resulting in 93,568 individuals born 1925-1967 (Figure 2), >50,000 of which participated ≥3 times.

The Age 40 Program Oslo
The study was originally organized as a cardiovascular disease (CVD) risk screening by the Oslo municipality health council in collaboration with the National Health Screening Service in Norway [4] The screening included anthropometric measurements, blood pressure measurements, non-fasting blood samples and a questionnaire, with all Oslo residents invited between 1981 and 1988 the year they turned 40 years of age (participation 55%). [5]A similar, restructured screening did continue in some districts of Oslo until 1999, but the quality of the data collected after 1988 was reported to have poor quality. [6]n total (1981-1999), 20,740 women and 18,754 men (from 104,482 invited) participated in the Age 40 Program Oslo. [7]NCDNOR includes 37,155 participants of the Age 40 Program Oslo (Figure 1).We excluded participants with unreliable age or inclusion date (n=69), and participants included after 1988 (n=13,200), resulting in 23,886 individuals born 1934-1949 (Figure 2).

The Age 40 Program
Inspired by the Age 40 Program in Oslo, the National Health Screening Service in Norway initiated cardiovascular disease risk screening programs in several other counties between 1985 and 1987, which also including anthropometric measurements, blood pressure measurements, non-fasting blood samples and a questionnaire, and by 1993 all Norwegian counties had conducted at least one screening. [8]All individuals residing in the county and aged 40-42 years by December 31 st the year the study started in that county were invited. [9]Some counties also invited residents aged <40 and 43-44 years, and between 1985 and 1993, three counties also invited those aged 65-67 years. [10]ome individuals moved between counties and were invited more than once (exact number unknown), resulting in ̴ 10,000 individuals participating several times.The overall participation was ~69%, with higher rates in the beginning (e.g.Østfold county: 81% in 1985 vs. 52% in 1999). [11]he measurement protocols of the Age 40 Program remained relatively unchanged over time and was almost identical to the one used in the Norwegian Counties Study, [8] but the questionnaire was revised three times (1988, 1994, 1997) and grew from one to three pages.Some counties/municipalities included additional measurements and/or questionnaire items some of the years the study was conducted. [7]CDNOR includes 417,097 participants of the Age 40 Program (Figure 1).We excluded participants aged <20 years (n=153), resulting in 416,944 individuals born 1900-1974 (Figure 2).
The HUNT Study The HUNT Study is the largest running health study in Norway and has been conducted four times (1984-1986, 1995-1997, 2008, and 2018-2019 (Figure 1)).It is designed to cover a broad range of health-related topics via questionnaires, interviews, clinical examinations, laboratory measurements and storage of biological samples.Following the completion of the fourth wave of the HUNT Study, which for the first time included participants from both the (former) North-and South-Trøndelag counties, an updated cohort profile for the HUNT Study was published giving detailed descriptions and information on the different studys. [12]he participation ranges from 89% in 1984-86 to 43% in 2018-19 (HUNT4 in South-Trøndelag (HUNT4ST)). [12]Almost 250,000 individuals have participated in HUNT, totalling 380,000 person-observations (~40,000 having participated three or four times).NCDNOR includes 228,955 participants from the HUNT Study.We excluded participants aged <20 years (n=3,039) resulting in 225,916 individuals born 1882-1999 (Figure 2).More details about the HUNT study are provided through the HUNT research centre's web pages (https://www.ntnu.edu/hunt)and the HUNT databank (https://hunt-db.medisin.ntnu.no/hunt-db/)

The Tromsø Study
The Tromsø Study is the longest running health study in Norway, including seven studies (Tromsø1-Tromsø7: 1974, 1979-1980, 1986-1987, 1994-1995, 2001, 2007-2008, and  2015-2016).The study has evolved from a combined cardiovascular disease risk screening and research study in the 1970s to a multipurpose health study covering a broad range of topics with data collections including questionnaires, interviews, biological sampling, and clinical examinations.An updated cohort profile was published recently giving detailed descriptions and information on the different Tromsø studies. [13] ). [14] The studies used the same protocols for anthropometric measurements, blood pressure measurements and non-fasting blood, and all studies used about 50 core CONOR questions agreed upon before the first CONOR study in Tromsø in 1994. [15,16]Most of the individual studies also made several additional measurements and included additional questionnaire items. [16]The overall participation rate in CONOR was ~58% but varied between studies from ~30% (I-HUBRO) to Harmonized into variable with three categories (0 Never, 1 Former daily smoker, 2 Current daily smoker) using a combination of original summary variables ('Smoking status summarized') and separate dichotomous variables ('Smoking status separate variables').The syntaxes (Stata do-files) used to harmonize the data is available upon request to the corresponding author.Cigarettes/day Harmonized from original variables, which for all surveys except the Age 40 Program Oslo were numerical in their original form.In the Age 40 Program Oslo, the original variable was categorical (0, 1-4, 5-9, 10-14, 15-19, 20-24 and ≥25 cigarettes per day).For the present study we harmonized cigarettes per day from the Age 40 Program Oslo with data from the other surveys by creating a numerical variable from these categories using the following rules: 0=0, 1-4=2.x x (…) last 2 weeks (5 cat: none but not a teetotaller, 1-4 times, 5-10 times, >10 times, I never drink/teetotaller) x (…) beer/wine/spirits last week (4 cat: not, 1 time, 2-3 times, 4 or more) x (…) beer/wine/spirits consumption now (5 cat: never/few times a year, 1-2/month, 1/week, 2-3/week, every day) x x Frequency of alcohol consumption numerical Number of times/days per month drinking alcohol (≥9 times equivalent to ≥2 times/week) x x x x x x Number of units/glasses/bottles/drinks/grams drunken Number of units beer/wine/spirits per 2 weeks usually x x x x x x x x x Heavy drinking -How often … on the same occasion (…) do you drink ≥5 units of alcohol on the same occasion?(4 cat: never, monthly, weekly, daily) x (…) do you drink ≥5 units of alcohol (…)? (5 cat: not the past year, a few times, 1-2/month, 1-2/week, 3+/week) x (…) do you drink ≥6 units of alcohol (…)? (5 cat: never, less <1/month, monthly, weekly, daily or almost daily) x x x x Number of 24-hour periods with ≥5 glasses or drinks with alcohol last year (≥12 times/year) x x Alcohol problems Have you had periods in life when you drank too much, or at least a bit too much?(3 cat: no, in doubt, yes) x Have you, in ≥1 periods the last 5 years, consumed so much alcohol that it inhibited your work, social life or both?(yes)

The Norwegian Counties Study
The

x
Abbreviations: CONOR, Cohort of Norway; HUNT, the Trøndelag Health Study (4S refers to the HUNT4 in South-Trøndelag); NCS, the Norwegian Counties Study; A40P, the Age 40 Program.a Translated from Norwegian to English *The questionnaire in the Age 40 Program Oslo did not include items covering alcohol consumption **The questionnaire used in the first two rounds of data collection in the Norwegian Counties Study (1974-1983) and in the Age 40 Program (1985-1994) did not include items covering alcohol consumption Finnmark o All residents aged 35-52 on December 31 st 1977 (born 1925-1942) o Everyone invited to the 1 st study still residing in Finnmark aged 23-34 o An 11% random sample of residents not invited to the 1 st study aged 23-34 o A 10% random sample of residents aged 20-22 • Sogn og Fjordane o All residents aged 40-54 on December 31 st 1980 (born 1926-1940) o Everyone invited to the 10% random sample in the 1 st study still residing in Sogn og Fjordane aged 25-39 o An 11% random sample of residents not invited to the 1 st study aged 25-39 o A 10% random sample of residents aged 20-24 o A 50% random sample of residents aged 17-19 in three of the county's municipalities • Oppland o All residents aged 40-54 on December 31st 1981 (born 1927-1941)

smoking/Age when starting
Abbreviations: CONOR, Cohort of Norway; HUNT, the Trøndelag Health Study (4S refers to the HUNT4 in South-Trøndelag); NCS, the Norwegian Counties Study; A40P, the Age 40 Program; A40PO, the Age 40 Program Oslo.
* Translated from Norwegian to EnglishSupplementary Table S2b.Harmonization and data cleaning*

også Her skal du også regne med gang
5, 5-9=7, 10-14=12, 15-19=17, 20-24=22 and ≥25=27 cigarettes per day.The syntaxes (Stata do-files) used to harmonize the data is available upon request to the corresponding author.Smoking years Except for (HUNT4, HUNT4ST, the 4th Norwegian Counites study), Smoking years was harmonized using the original numerical variables.For HUNT4, HUNT4ST, the 4th Norwegian Counites study we calculated Smoking years from self-reported age of daily smoking initiation and cessation (or age at participation if current daily smoker).The syntaxes (Stata dofiles) used to harmonize the data is available upon request to the corresponding author.Most of the time you walk, and you often have to walk up stairs and lift various items.Examples include mail delivery and construction work.4 You have heavy physical work.You carry heavy burdens and carry out physically strenuous work, for example, work including digging and shovelling Bevegelse og kroppslig anstrengelse i deres fritid.Hvis aktiviteten varierer meget f.eks.mellom sommer og vinter så ta et gjennomsnitt.Spørsmålet gjelder bare det siste året.Sett kryss i ruten hvor 'JA' passer best.Bevegelse og kroppslig anstrengelse i deres fritid.Hvis aktiviteten varierer meget f.eks.mellom sommer og vinter så ta et gjennomsnitt.Spørsmålet gjelder bare det siste året.Sett kryss i ruta hvor 'JA' passer best.You walk around quite a bit at your workplace but do not have to carry heavy items-for example, light industrial work, non-sedentary office work, inspection and the like.('Walk+lift"): Most of the time you walk, and you often have to walk up stairs and lift various items.Examples include mail delivery and construction work.('Heavy labour"): You have heavy physical work.You carry heavy burdens and carry out physically strenuous work-for example, work including digging and shovelling The following section deals with your spare-time physical activity.The table outlines four different levels.Please read the table carefully and then check appropriate boxes below: 'Sedentary": Almost completely inactive: reading, TV watching, movies, etc. 'Low": Some physical activity during at least 4 hours per week: riding a bicycle or walking to work, walking or skiing with the family, gardening.'Moderate": Regular activity: such as heavy gardening, running, calisthenics, tennis, etc. 'High": Regular hard physical training for competition in running events, soccer, racing, European handball, etc. several times per week.Answered if Exercise frequency = 3, 4 or 5. **Participants could only choose one category.

Table S7 . Participant characteristics by cluster of noncommunicable disease risk factors (n=625,364) a
Abbreviations: No., number; IQR, inter-quartile range a Data are presented as number of participants or column percentages, unless otherwise indicated.Because of rounding, percentages may not sum up to 100%.bAtfirst participation c For current and former smokers Supplementary TableS8.

Participant characteristics at trajectory entry by latent class of smoking intensity trajectory (n=22,412) a
d At trajectory entry e For current and former smokers Supplementary

alcohol consumption (grams/14 days)
Participant characteristics are described at trajectory entry, ie the first participation in the period 30-40 years of age.
b c At trajectory entry d For current and former smoker Supplementary

alcohol consumption (grams/14 days)
Participant characteristics are described at trajectory entry, ie the first participation in the period 30-40 years of age.
b c At trajectory entry d For current and former smokers Supplementary

Body mass index (kg/m 2 )
Abbreviations: No., number; IQR, inter-quartile range a Data are presented as number of participants or column percentages, unless otherwise indicated.Because of rounding, percentages may not sum up to 100%.
b Participant characteristics are described at trajectory entry, ie the first participation in the period 30-40 years of age.cAt trajectory entry d For current and former smokers Supplementary TableS12.

Body mass index (kg/m 2 )
Abbreviations: No., number; IQR, inter-quartile range a Data are presented as number of participants or column percentages, unless otherwise indicated.Because of rounding, percentages may not sum up to 100%.b Participant characteristics are described at trajectory entry, i.e. the first participation in the period 30-40 years of age.
c At trajectory entry d For current and former smokers